categories
string | doi
string | id
string | year
float64 | venue
string | link
string | updated
string | published
string | title
string | abstract
string | authors
sequence |
|---|---|---|---|---|---|---|---|---|---|---|
null | null | 2406.13129 | null | null | http://arxiv.org/pdf/2406.13129v1 | 2024-06-19T00:46:48Z | 2024-06-19T00:46:48Z | M3T: Multi-Modal Medical Transformer to bridge Clinical Context with
Visual Insights for Retinal Image Medical Description Generation | Automated retinal image medical description generation is crucial for streamlining medical diagnosis and treatment planning. Existing challenges include the reliance on learned retinal image representations, difficulties in handling multiple imaging modalities, and the lack of clinical context in visual representations. Addressing these issues, we propose the Multi-Modal Medical Transformer (M3T), a novel deep learning architecture that integrates visual representations with diagnostic keywords. Unlike previous studies focusing on specific aspects, our approach efficiently learns contextual information and semantics from both modalities, enabling the generation of precise and coherent medical descriptions for retinal images. Experimental studies on the DeepEyeNet dataset validate the success of M3T in meeting ophthalmologists' standards, demonstrating a substantial 13.5% improvement in BLEU@4 over the best-performing baseline model. | [
"['Nagur Shareef Shaik' 'Teja Krishna Cherukuri' 'Dong Hye Ye']"
] |
null | null | 2406.13130 | null | null | http://arxiv.org/pdf/2406.13130v1 | 2024-06-19T00:47:38Z | 2024-06-19T00:47:38Z | Advancing Retail Data Science: Comprehensive Evaluation of Synthetic
Data | The evaluation of synthetic data generation is crucial, especially in the retail sector where data accuracy is paramount. This paper introduces a comprehensive framework for assessing synthetic retail data, focusing on fidelity, utility, and privacy. Our approach differentiates between continuous and discrete data attributes, providing precise evaluation criteria. Fidelity is measured through stability and generalizability. Stability ensures synthetic data accurately replicates known data distributions, while generalizability confirms its robustness in novel scenarios. Utility is demonstrated through the synthetic data's effectiveness in critical retail tasks such as demand forecasting and dynamic pricing, proving its value in predictive analytics and strategic planning. Privacy is safeguarded using Differential Privacy, ensuring synthetic data maintains a perfect balance between resembling training and holdout datasets without compromising security. Our findings validate that this framework provides reliable and scalable evaluation for synthetic retail data. It ensures high fidelity, utility, and privacy, making it an essential tool for advancing retail data science. This framework meets the evolving needs of the retail industry with precision and confidence, paving the way for future advancements in synthetic data methodologies. | [
"['Yu Xia' 'Chi-Hua Wang' 'Joshua Mabry' 'Guang Cheng']"
] |
null | null | 2406.13133 | null | null | http://arxiv.org/pdf/2406.13133v1 | 2024-06-19T00:53:48Z | 2024-06-19T00:53:48Z | PathoLM: Identifying pathogenicity from the DNA sequence through the
Genome Foundation Model | Pathogen identification is pivotal in diagnosing, treating, and preventing diseases, crucial for controlling infections and safeguarding public health. Traditional alignment-based methods, though widely used, are computationally intense and reliant on extensive reference databases, often failing to detect novel pathogens due to their low sensitivity and specificity. Similarly, conventional machine learning techniques, while promising, require large annotated datasets and extensive feature engineering and are prone to overfitting. Addressing these challenges, we introduce PathoLM, a cutting-edge pathogen language model optimized for the identification of pathogenicity in bacterial and viral sequences. Leveraging the strengths of pre-trained DNA models such as the Nucleotide Transformer, PathoLM requires minimal data for fine-tuning, thereby enhancing pathogen detection capabilities. It effectively captures a broader genomic context, significantly improving the identification of novel and divergent pathogens. We developed a comprehensive data set comprising approximately 30 species of viruses and bacteria, including ESKAPEE pathogens, seven notably virulent bacterial strains resistant to antibiotics. Additionally, we curated a species classification dataset centered specifically on the ESKAPEE group. In comparative assessments, PathoLM dramatically outperforms existing models like DciPatho, demonstrating robust zero-shot and few-shot capabilities. Furthermore, we expanded PathoLM-Sp for ESKAPEE species classification, where it showed superior performance compared to other advanced deep learning methods, despite the complexities of the task. | [
"['Sajib Acharjee Dip' 'Uddip Acharjee Shuvo' 'Tran Chau' 'Haoqiu Song'\n 'Petra Choi' 'Xuan Wang' 'Liqing Zhang']"
] |
null | null | 2406.13137 | null | null | http://arxiv.org/pdf/2406.13137v1 | 2024-06-19T01:03:23Z | 2024-06-19T01:03:23Z | Efficient Sharpness-Aware Minimization for Molecular Graph Transformer
Models | Sharpness-aware minimization (SAM) has received increasing attention in computer vision since it can effectively eliminate the sharp local minima from the training trajectory and mitigate generalization degradation. However, SAM requires two sequential gradient computations during the optimization of each step: one to obtain the perturbation gradient and the other to obtain the updating gradient. Compared with the base optimizer (e.g., Adam), SAM doubles the time overhead due to the additional perturbation gradient. By dissecting the theory of SAM and observing the training gradient of the molecular graph transformer, we propose a new algorithm named GraphSAM, which reduces the training cost of SAM and improves the generalization performance of graph transformer models. There are two key factors that contribute to this result: (i) textit{gradient approximation}: we use the updating gradient of the previous step to approximate the perturbation gradient at the intermediate steps smoothly (textbf{increases efficiency}); (ii) textit{loss landscape approximation}: we theoretically prove that the loss landscape of GraphSAM is limited to a small range centered on the expected loss of SAM (textbf{guarantees generalization performance}). The extensive experiments on six datasets with different tasks demonstrate the superiority of GraphSAM, especially in optimizing the model update process. The code is in:https://github.com/YL-wang/GraphSAM/tree/graphsam | [
"['Yili Wang' 'Kaixiong Zhou' 'Ninghao Liu' 'Ying Wang' 'Xin Wang']"
] |
null | null | 2406.13145 | null | null | http://arxiv.org/pdf/2406.13145v1 | 2024-06-19T01:45:18Z | 2024-06-19T01:45:18Z | Constructing and Evaluating Digital Twins: An Intelligent Framework for
DT Development | The development of Digital Twins (DTs) represents a transformative advance for simulating and optimizing complex systems in a controlled digital space. Despite their potential, the challenge of constructing DTs that accurately replicate and predict the dynamics of real-world systems remains substantial. This paper introduces an intelligent framework for the construction and evaluation of DTs, specifically designed to enhance the accuracy and utility of DTs in testing algorithmic performance. We propose a novel construction methodology that integrates deep learning-based policy gradient techniques to dynamically tune the DT parameters, ensuring high fidelity in the digital replication of physical systems. Moreover, the Mean STate Error (MSTE) is proposed as a robust metric for evaluating the performance of algorithms within these digital space. The efficacy of our framework is demonstrated through extensive simulations that show our DT not only accurately mirrors the physical reality but also provides a reliable platform for algorithm evaluation. This work lays a foundation for future research into DT technologies, highlighting pathways for both theoretical enhancements and practical implementations in various industries. | [
"['Longfei Ma' 'Nan Cheng' 'Xiucheng Wang' 'Jiong Chen' 'Yinjun Gao'\n 'Dongxiao Zhang' 'Jun-Jie Zhang']"
] |
null | null | 2406.13151 | null | null | http://arxiv.org/pdf/2406.13151v1 | 2024-06-19T01:57:21Z | 2024-06-19T01:57:21Z | von Mises Quasi-Processes for Bayesian Circular Regression | The need for regression models to predict circular values arises in many scientific fields. In this work we explore a family of expressive and interpretable distributions over circle-valued random functions related to Gaussian processes targeting two Euclidean dimensions conditioned on the unit circle. The resulting probability model has connections with continuous spin models in statistical physics. Moreover, its density is very simple and has maximum-entropy, unlike previous Gaussian process-based approaches, which use wrapping or radial marginalization. For posterior inference, we introduce a new Stratonovich-like augmentation that lends itself to fast Markov Chain Monte Carlo sampling. We argue that transductive learning in these models favors a Bayesian approach to the parameters. We present experiments applying this model to the prediction of (i) wind directions and (ii) the percentage of the running gait cycle as a function of joint angles. | [
"['Yarden Cohen' 'Alexandre Khae Wu Navarro' 'Jes Frellsen'\n 'Richard E. Turner' 'Raziel Riemer' 'Ari Pakman']"
] |
null | null | 2406.13154 | null | null | http://arxiv.org/pdf/2406.13154v2 | 2024-06-21T19:01:31Z | 2024-06-19T02:09:15Z | Conditional score-based diffusion models for solving inverse problems in
mechanics | We propose a framework to perform Bayesian inference using conditional score-based diffusion models to solve a class of inverse problems in mechanics involving the inference of a specimen's spatially varying material properties from noisy measurements of its mechanical response to loading. Conditional score-based diffusion models are generative models that learn to approximate the score function of a conditional distribution using samples from the joint distribution. More specifically, the score functions corresponding to multiple realizations of the measurement are approximated using a single neural network, the so-called score network, which is subsequently used to sample the posterior distribution using an appropriate Markov chain Monte Carlo scheme based on Langevin dynamics. Training the score network only requires simulating the forward model. Hence, the proposed approach can accommodate black-box forward models and complex measurement noise. Moreover, once the score network has been trained, it can be re-used to solve the inverse problem for different realizations of the measurements. We demonstrate the efficacy of the proposed approach on a suite of high-dimensional inverse problems in mechanics that involve inferring heterogeneous material properties from noisy measurements. Some examples we consider involve synthetic data, while others include data collected from actual elastography experiments. Further, our applications demonstrate that the proposed approach can handle different measurement modalities, complex patterns in the inferred quantities, non-Gaussian and non-additive noise models, and nonlinear black-box forward models. The results show that the proposed framework can solve large-scale physics-based inverse problems efficiently. | [
"['Agnimitra Dasgupta' 'Harisankar Ramaswamy' 'Javier Murgoitio Esandi'\n 'Ken Foo' 'Runze Li' 'Qifa Zhou' 'Brendan Kennedy' 'Assad Oberai']"
] |
null | null | 2406.13161 | null | null | http://arxiv.org/pdf/2406.13161v1 | 2024-06-19T02:29:59Z | 2024-06-19T02:29:59Z | APPL: A Prompt Programming Language for Harmonious Integration of
Programs and Large Language Model Prompts | Large Language Models (LLMs) have become increasingly capable of handling diverse tasks with the aid of well-crafted prompts and integration of external tools, but as task complexity rises, the workflow involving LLMs can be complicated and thus challenging to implement and maintain. To address this challenge, we propose APPL, A Prompt Programming Language that acts as a bridge between computer programs and LLMs, allowing seamless embedding of prompts into Python functions, and vice versa. APPL provides an intuitive and Python-native syntax, an efficient parallelized runtime with asynchronous semantics, and a tracing module supporting effective failure diagnosis and replaying without extra costs. We demonstrate that APPL programs are intuitive, concise, and efficient through three representative scenarios: Chain-of-Thought with self-consistency (CoT-SC), ReAct tool use agent, and multi-agent chat. Experiments on three parallelizable workflows further show that APPL can effectively parallelize independent LLM calls, with a significant speedup ratio that almost matches the estimation. | [
"['Honghua Dong' 'Qidong Su' 'Yubo Gao' 'Zhaoyu Li' 'Yangjun Ruan'\n 'Gennady Pekhimenko' 'Chris J. Maddison' 'Xujie Si']"
] |
null | null | 2406.13162 | null | null | http://arxiv.org/pdf/2406.13162v1 | 2024-06-19T02:31:23Z | 2024-06-19T02:31:23Z | AntibodyFlow: Normalizing Flow Model for Designing Antibody
Complementarity-Determining Regions | Therapeutic antibodies have been extensively studied in drug discovery and development in the past decades. Antibodies are specialized protective proteins that bind to antigens in a lock-to-key manner. The binding strength/affinity between an antibody and a specific antigen is heavily determined by the complementarity-determining regions (CDRs) on the antibodies. Existing machine learning methods cast in silico development of CDRs as either sequence or 3D graph (with a single chain) generation tasks and have achieved initial success. However, with CDR loops having specific geometry shapes, learning the 3D geometric structures of CDRs remains a challenge. To address this issue, we propose AntibodyFlow, a 3D flow model to design antibody CDR loops. Specifically, AntibodyFlow first constructs the distance matrix, then predicts amino acids conditioned on the distance matrix. Also, AntibodyFlow conducts constraint learning and constrained generation to ensure valid 3D structures. Experimental results indicate that AntibodyFlow outperforms the best baseline consistently with up to 16.0% relative improvement in validity rate and 24.3% relative reduction in geometric graph level error (root mean square deviation, RMSD). | [
"['Bohao Xu' 'Yanbo Wang' 'Wenyu Chen' 'Shimin Shan']"
] |
null | null | 2406.13166 | null | null | http://arxiv.org/pdf/2406.13166v1 | 2024-06-19T02:45:32Z | 2024-06-19T02:45:32Z | Enhancing supply chain security with automated machine learning | This study tackles the complexities of global supply chains, which are increasingly vulnerable to disruptions caused by port congestion, material shortages, and inflation. To address these challenges, we explore the application of machine learning methods, which excel in predicting and optimizing solutions based on large datasets. Our focus is on enhancing supply chain security through fraud detection, maintenance prediction, and material backorder forecasting. We introduce an automated machine learning framework that streamlines data analysis, model construction, and hyperparameter optimization for these tasks. By automating these processes, our framework improves the efficiency and effectiveness of supply chain security measures. Our research identifies key factors that influence machine learning performance, including sampling methods, categorical encoding, feature selection, and hyperparameter optimization. We demonstrate the importance of considering these factors when applying machine learning to supply chain challenges. Traditional mathematical programming models often struggle to cope with the complexity of large-scale supply chain problems. Our study shows that machine learning methods can provide a viable alternative, particularly when dealing with extensive datasets and complex patterns. The automated machine learning framework presented in this study offers a novel approach to supply chain security, contributing to the existing body of knowledge in the field. Its comprehensive automation of machine learning processes makes it a valuable contribution to the domain of supply chain management. | [
"['Haibo Wang' 'Lutfu S. Sua' 'Bahram Alidaee']"
] |
null | null | 2406.13173 | null | null | http://arxiv.org/pdf/2406.13173v2 | 2024-06-30T01:22:09Z | 2024-06-19T03:07:33Z | Biomedical Visual Instruction Tuning with Clinician Preference Alignment | Recent advancements in multimodal foundation models have showcased impressive capabilities in understanding and reasoning with visual and textual information. Adapting these foundation models trained for general usage to specialized domains like biomedicine requires large-scale domain-specific instruction datasets. While existing works have explored curating such datasets automatically, the resultant datasets are not explicitly aligned with domain expertise. In this work, we propose a data-centric framework, Biomedical Visual Instruction Tuning with Clinician Preference Alignment (BioMed-VITAL), that incorporates clinician preferences into both stages of generating and selecting instruction data for tuning biomedical multimodal foundation models. First, during the generation stage, we prompt the GPT-4V generator with a diverse set of clinician-selected demonstrations for preference-aligned data candidate generation. Then, during the selection phase, we train a separate selection model, which explicitly distills clinician and policy-guided model preferences into a rating function to select high-quality data for medical instruction tuning. Results show that the model tuned with the instruction-following data from our method demonstrates a significant improvement in open visual chat (18.5% relatively) and medical VQA (win rate up to 81.73%). Our instruction-following data and models are available at BioMed-VITAL.github.io. | [
"['Hejie Cui' 'Lingjun Mao' 'Xin Liang' 'Jieyu Zhang' 'Hui Ren'\n 'Quanzheng Li' 'Xiang Li' 'Carl Yang']"
] |
null | null | 2406.13175 | null | null | http://arxiv.org/pdf/2406.13175v1 | 2024-06-19T03:13:11Z | 2024-06-19T03:13:11Z | Sparse High Rank Adapters | Low Rank Adaptation (LoRA) has gained massive attention in the recent generative AI research. One of the main advantages of LoRA is its ability to be fused with pretrained models adding no overhead during inference. However, from a mobile deployment standpoint, we can either avoid inference overhead in the fused mode but lose the ability to switch adapters rapidly, or suffer significant (up to 30% higher) inference latency while enabling rapid switching in the unfused mode. LoRA also exhibits concept-loss when multiple adapters are used concurrently. In this paper, we propose Sparse High Rank Adapters (SHiRA), a new paradigm which incurs no inference overhead, enables rapid switching, and significantly reduces concept-loss. Specifically, SHiRA can be trained by directly tuning only 1-2% of the base model weights while leaving others unchanged. This results in a highly sparse adapter which can be switched directly in the fused mode. We further provide theoretical and empirical insights on how high sparsity in SHiRA can aid multi-adapter fusion by reducing concept loss. Our extensive experiments on LVMs and LLMs demonstrate that finetuning only a small fraction of the parameters in the base model is sufficient for many tasks while enabling both rapid switching and multi-adapter fusion. Finally, we provide a latency- and memory-efficient SHiRA implementation based on Parameter-Efficient Finetuning (PEFT) Library. This implementation trains at nearly the same speed as LoRA while consuming lower peak GPU memory, thus making SHiRA easy to adopt for practical use cases. | [
"['Kartikeya Bhardwaj' 'Nilesh Prasad Pandey' 'Sweta Priyadarshi'\n 'Viswanath Ganapathy' 'Rafael Esteves' 'Shreya Kadambi'\n 'Shubhankar Borse' 'Paul Whatmough' 'Risheek Garrepalli'\n 'Mart Van Baalen' 'Harris Teague' 'Markus Nagel']"
] |
null | null | 2406.13183 | null | null | http://arxiv.org/pdf/2406.13183v1 | 2024-06-19T03:29:51Z | 2024-06-19T03:29:51Z | Communication-Efficient and Privacy-Preserving Decentralized
Meta-Learning | Distributed learning, which does not require gathering training data in a central location, has become increasingly important in the big-data era. In particular, random-walk-based decentralized algorithms are flexible in that they do not need a central server trusted by all clients and do not require all clients to be active in all iterations. However, existing distributed learning algorithms assume that all learning clients share the same task. In this paper, we consider the more difficult meta-learning setting, in which different clients perform different (but related) tasks with limited training data. To reduce communication cost and allow better privacy protection, we propose LoDMeta (Local Decentralized Meta-learning) with the use of local auxiliary optimization parameters and random perturbations on the model parameter. Theoretical results are provided on both convergence and privacy analysis. Empirical results on a number of few-shot learning data sets demonstrate that LoDMeta has similar meta-learning accuracy as centralized meta-learning algorithms, but does not require gathering data from each client and is able to better protect data privacy for each client. | [
"['Hansi Yang' 'James T. Kwok']"
] |
null | null | 2406.13187 | null | null | http://arxiv.org/pdf/2406.13187v1 | 2024-06-19T03:35:26Z | 2024-06-19T03:35:26Z | Boosting Consistency in Dual Training for Long-Tailed Semi-Supervised
Learning | While long-tailed semi-supervised learning (LTSSL) has received tremendous attention in many real-world classification problems, existing LTSSL algorithms typically assume that the class distributions of labeled and unlabeled data are almost identical. Those LTSSL algorithms built upon the assumption can severely suffer when the class distributions of labeled and unlabeled data are mismatched since they utilize biased pseudo-labels from the model. To alleviate this problem, we propose a new simple method that can effectively utilize unlabeled data from unknown class distributions through Boosting cOnsistency in duAl Training (BOAT). Specifically, we construct the standard and balanced branch to ensure the performance of the head and tail classes, respectively. Throughout the training process, the two branches incrementally converge and interact with each other, eventually resulting in commendable performance across all classes. Despite its simplicity, we show that BOAT achieves state-of-the-art performance on a variety of standard LTSSL benchmarks, e.g., an averaged 2.7% absolute increase in test accuracy against existing algorithms when the class distributions of labeled and unlabeled data are mismatched. Even when the class distributions are identical, BOAT consistently outperforms many sophisticated LTSSL algorithms. We carry out extensive ablation studies to tease apart the factors that are the most important to the success of BOAT. The source code is available at https://github.com/Gank0078/BOAT. | [
"['Kai Gan' 'Tong Wei' 'Min-Ling Zhang']"
] |
null | null | 2406.13188 | null | null | http://arxiv.org/pdf/2406.13188v1 | 2024-06-19T03:37:52Z | 2024-06-19T03:37:52Z | Synthetic Context Generation for Question Generation | Despite rapid advancements in large language models (LLMs), QG remains a challenging problem due to its complicated process, open-ended nature, and the diverse settings in which question generation occurs. A common approach to address these challenges involves fine-tuning smaller, custom models using datasets containing background context, question, and answer. However, obtaining suitable domain-specific datasets with appropriate context is often more difficult than acquiring question-answer pairs. In this paper, we investigate training QG models using synthetic contexts generated by LLMs from readily available question-answer pairs. We conduct a comprehensive study to answer critical research questions related to the performance of models trained on synthetic contexts and their potential impact on QG research and applications. Our empirical results reveal: 1) contexts are essential for QG tasks, even if they are synthetic; 2) fine-tuning smaller language models has the capability of achieving better performances as compared to prompting larger language models; and 3) synthetic context and real context could achieve comparable performances. These findings highlight the effectiveness of synthetic contexts in QG and paves the way for future advancements in the field. | [
"['Naiming Liu' 'Zichao Wang' 'Richard Baraniuk']"
] |
null | null | 2406.13193 | null | null | http://arxiv.org/pdf/2406.13193v1 | 2024-06-19T03:59:46Z | 2024-06-19T03:59:46Z | PRESTO: Progressive Pretraining Enhances Synthetic Chemistry Outcomes | Multimodal Large Language Models (MLLMs) have seen growing adoption across various scientific disciplines. These advancements encourage the investigation of molecule-text modeling within synthetic chemistry, a field dedicated to designing and conducting chemical reactions to synthesize new compounds with desired properties and applications. Current approaches, however, often neglect the critical role of multiple molecule graph interaction in understanding chemical reactions, leading to suboptimal performance in synthetic chemistry tasks. This study introduces PRESTO(Progressive Pretraining Enhances Synthetic Chemistry Outcomes), a new framework that bridges the molecule-text modality gap by integrating a comprehensive benchmark of pretraining strategies and dataset configurations. It progressively improves multimodal LLMs through cross-modal alignment and multi-graph understanding. Our extensive experiments demonstrate that PRESTO offers competitive results in downstream synthetic chemistry tasks. The code can be found at https://github.com/IDEA-XL/PRESTO. | [
"['He Cao' 'Yanjun Shao' 'Zhiyuan Liu' 'Zijing Liu' 'Xiangru Tang'\n 'Yuan Yao' 'Yu Li']"
] |
null | null | 2406.13200 | null | null | http://arxiv.org/pdf/2406.13200v1 | 2024-06-19T04:14:57Z | 2024-06-19T04:14:57Z | RobGC: Towards Robust Graph Condensation | Graph neural networks (GNNs) have attracted widespread attention for their impressive capability of graph representation learning. However, the increasing prevalence of large-scale graphs presents a significant challenge for GNN training due to their computational demands, limiting the applicability of GNNs in various scenarios. In response to this challenge, graph condensation (GC) is proposed as a promising acceleration solution, focusing on generating an informative compact graph that enables efficient training of GNNs while retaining performance. Despite the potential to accelerate GNN training, existing GC methods overlook the quality of large training graphs during both the training and inference stages. They indiscriminately emulate the training graph distributions, making the condensed graphs susceptible to noises within the training graph and significantly impeding the application of GC in intricate real-world scenarios. To address this issue, we propose robust graph condensation (RobGC), a plug-and-play approach for GC to extend the robustness and applicability of condensed graphs in noisy graph structure environments. Specifically, RobGC leverages the condensed graph as a feedback signal to guide the denoising process on the original training graph. A label propagation-based alternating optimization strategy is in place for the condensation and denoising processes, contributing to the mutual purification of the condensed graph and training graph. Additionally, as a GC method designed for inductive graph inference, RobGC facilitates test-time graph denoising by leveraging the noise-free condensed graph to calibrate the structure of the test graph. Extensive experiments show that RobGC is compatible with various GC methods, significantly boosting their robustness under different types and levels of graph structural noises. | [
"['Xinyi Gao' 'Hongzhi Yin' 'Tong Chen' 'Guanhua Ye' 'Wentao Zhang'\n 'Bin Cui']"
] |
null | null | 2406.13201 | null | null | http://arxiv.org/pdf/2406.13201v1 | 2024-06-19T04:20:12Z | 2024-06-19T04:20:12Z | Toward Structure Fairness in Dynamic Graph Embedding: A Trend-aware Dual
Debiasing Approach | Recent studies successfully learned static graph embeddings that are structurally fair by preventing the effectiveness disparity of high- and low-degree vertex groups in downstream graph mining tasks. However, achieving structure fairness in dynamic graph embedding remains an open problem. Neglecting degree changes in dynamic graphs will significantly impair embedding effectiveness without notably improving structure fairness. This is because the embedding performance of high-degree and low-to-high-degree vertices will significantly drop close to the generally poorer embedding performance of most slightly changed vertices in the long-tail part of the power-law distribution. We first identify biased structural evolutions in a dynamic graph based on the evolving trend of vertex degree and then propose FairDGE, the first structurally Fair Dynamic Graph Embedding algorithm. FairDGE learns biased structural evolutions by jointly embedding the connection changes among vertices and the long-short-term evolutionary trend of vertex degrees. Furthermore, a novel dual debiasing approach is devised to encode fair embeddings contrastively, customizing debiasing strategies for different biased structural evolutions. This innovative debiasing strategy breaks the effectiveness bottleneck of embeddings without notable fairness loss. Extensive experiments demonstrate that FairDGE achieves simultaneous improvement in the effectiveness and fairness of embeddings. | [
"['Yicong Li' 'Yu Yang' 'Jiannong Cao' 'Shuaiqi Liu' 'Haoran Tang'\n 'Guandong Xu']"
] |
null | null | 2406.13214 | null | null | http://arxiv.org/pdf/2406.13214v1 | 2024-06-19T04:55:34Z | 2024-06-19T04:55:34Z | Self-Explainable Temporal Graph Networks based on Graph Information
Bottleneck | Temporal Graph Neural Networks (TGNN) have the ability to capture both the graph topology and dynamic dependencies of interactions within a graph over time. There has been a growing need to explain the predictions of TGNN models due to the difficulty in identifying how past events influence their predictions. Since the explanation model for a static graph cannot be readily applied to temporal graphs due to its inability to capture temporal dependencies, recent studies proposed explanation models for temporal graphs. However, existing explanation models for temporal graphs rely on post-hoc explanations, requiring separate models for prediction and explanation, which is limited in two aspects: efficiency and accuracy of explanation. In this work, we propose a novel built-in explanation framework for temporal graphs, called Self-Explainable Temporal Graph Networks based on Graph Information Bottleneck (TGIB). TGIB provides explanations for event occurrences by introducing stochasticity in each temporal event based on the Information Bottleneck theory. Experimental results demonstrate the superiority of TGIB in terms of both the link prediction performance and explainability compared to state-of-the-art methods. This is the first work that simultaneously performs prediction and explanation for temporal graphs in an end-to-end manner. | [
"['Sangwoo Seo' 'Sungwon Kim' 'Jihyeong Jung' 'Yoonho Lee' 'Chanyoung Park']"
] |
null | null | 2406.13216 | null | null | http://arxiv.org/pdf/2406.13216v1 | 2024-06-19T04:57:35Z | 2024-06-19T04:57:35Z | Combining Optimal Transport and Embedding-Based Approaches for More
Expressiveness in Unsupervised Graph Alignment | Unsupervised graph alignment finds the one-to-one node correspondence between a pair of attributed graphs by only exploiting graph structure and node features. One category of existing works first computes the node representation and then matches nodes with close embeddings, which is intuitive but lacks a clear objective tailored for graph alignment in the unsupervised setting. The other category reduces the problem to optimal transport (OT) via Gromov-Wasserstein (GW) learning with a well-defined objective but leaves a large room for exploring the design of transport cost. We propose a principled approach to combine their advantages motivated by theoretical analysis of model expressiveness. By noticing the limitation of discriminative power in separating matched and unmatched node pairs, we improve the cost design of GW learning with feature transformation, which enables feature interaction across dimensions. Besides, we propose a simple yet effective embedding-based heuristic inspired by the Weisfeiler-Lehman test and add its prior knowledge to OT for more expressiveness when handling non-Euclidean data. Moreover, we are the first to guarantee the one-to-one matching constraint by reducing the problem to maximum weight matching. The algorithm design effectively combines our OT and embedding-based predictions via stacking, an ensemble learning strategy. We propose a model framework named texttt{CombAlign} integrating all the above modules to refine node alignment progressively. Through extensive experiments, we demonstrate significant improvements in alignment accuracy compared to state-of-the-art approaches and validate the effectiveness of the proposed modules. | [
"['Songyang Chen' 'Yu Liu' 'Lei Zou' 'Zexuan Wang' 'Youfang Lin'\n 'Yuxing Chen' 'Anqun Pan']"
] |
null | null | 2406.13225 | null | null | http://arxiv.org/pdf/2406.13225v1 | 2024-06-19T05:26:02Z | 2024-06-19T05:26:02Z | Communication-Efficient Federated Knowledge Graph Embedding with
Entity-Wise Top-K Sparsification | Federated Knowledge Graphs Embedding learning (FKGE) encounters challenges in communication efficiency stemming from the considerable size of parameters and extensive communication rounds. However, existing FKGE methods only focus on reducing communication rounds by conducting multiple rounds of local training in each communication round, and ignore reducing the size of parameters transmitted within each communication round. To tackle the problem, we first find that universal reduction in embedding precision across all entities during compression can significantly impede convergence speed, underscoring the importance of maintaining embedding precision. We then propose bidirectional communication-efficient FedS based on Entity-Wise Top-K Sparsification strategy. During upload, clients dynamically identify and upload only the Top-K entity embeddings with the greater changes to the server. During download, the server first performs personalized embedding aggregation for each client. It then identifies and transmits the Top-K aggregated embeddings to each client. Besides, an Intermittent Synchronization Mechanism is used by FedS to mitigate negative effect of embedding inconsistency among shared entities of clients caused by heterogeneity of Federated Knowledge Graph. Extensive experiments across three datasets showcase that FedS significantly enhances communication efficiency with negligible (even no) performance degradation. | [
"['Xiaoxiong Zhang' 'Zhiwei Zeng' 'Xin Zhou' 'Dusit Niyato' 'Zhiqi Shen']"
] |
null | null | 2406.13228 | null | null | http://arxiv.org/pdf/2406.13228v1 | 2024-06-19T05:29:20Z | 2024-06-19T05:29:20Z | AGSOA:Graph Neural Network Targeted Attack Based on Average Gradient and
Structure Optimization | Graph Neural Networks(GNNs) are vulnerable to adversarial attack that cause performance degradation by adding small perturbations to the graph. Gradient-based attacks are one of the most commonly used methods and have achieved good performance in many attack scenarios. However, current gradient attacks face the problems of easy to fall into local optima and poor attack invisibility. Specifically, most gradient attacks use greedy strategies to generate perturbations, which tend to fall into local optima leading to underperformance of the attack. In addition, many attacks only consider the effectiveness of the attack and ignore the invisibility of the attack, making the attacks easily exposed leading to failure. To address the above problems, this paper proposes an attack on GNNs, called AGSOA, which consists of an average gradient calculation and a structre optimization module. In the average gradient calculation module, we compute the average of the gradient information over all moments to guide the attack to generate perturbed edges, which stabilizes the direction of the attack update and gets rid of undesirable local maxima. In the structure optimization module, we calculate the similarity and homogeneity of the target node's with other nodes to adjust the graph structure so as to improve the invisibility and transferability of the attack. Extensive experiments on three commonly used datasets show that AGSOA improves the misclassification rate by 2$%$-8$%$ compared to other state-of-the-art models. | [
"['Yang Chen' 'Bin Zhou']"
] |
null | null | 2406.13229 | null | null | http://arxiv.org/pdf/2406.13229v1 | 2024-06-19T05:31:59Z | 2024-06-19T05:31:59Z | Probing the Emergence of Cross-lingual Alignment during LLM Training | Multilingual Large Language Models (LLMs) achieve remarkable levels of zero-shot cross-lingual transfer performance. We speculate that this is predicated on their ability to align languages without explicit supervision from parallel sentences. While representations of translationally equivalent sentences in different languages are known to be similar after convergence, however, it remains unclear how such cross-lingual alignment emerges during pre-training of LLMs. Our study leverages intrinsic probing techniques, which identify which subsets of neurons encode linguistic features, to correlate the degree of cross-lingual neuron overlap with the zero-shot cross-lingual transfer performance for a given model. In particular, we rely on checkpoints of BLOOM, a multilingual autoregressive LLM, across different training steps and model scales. We observe a high correlation between neuron overlap and downstream performance, which supports our hypothesis on the conditions leading to effective cross-lingual transfer. Interestingly, we also detect a degradation of both implicit alignment and multilingual abilities in certain phases of the pre-training process, providing new insights into the multilingual pretraining dynamics. | [
"['Hetong Wang' 'Pasquale Minervini' 'Edoardo M. Ponti']"
] |
null | null | 2406.13232 | null | null | http://arxiv.org/pdf/2406.13232v1 | 2024-06-19T05:43:02Z | 2024-06-19T05:43:02Z | Towards Robust Evaluation: A Comprehensive Taxonomy of Datasets and
Metrics for Open Domain Question Answering in the Era of Large Language
Models | Open Domain Question Answering (ODQA) within natural language processing involves building systems that answer factual questions using large-scale knowledge corpora. Recent advances stem from the confluence of several factors, such as large-scale training datasets, deep learning techniques, and the rise of large language models. High-quality datasets are used to train models on realistic scenarios and enable the evaluation of the system on potentially unseen data. Standardized metrics facilitate comparisons between different ODQA systems, allowing researchers to objectively track advancements in the field. Our study presents a thorough examination of the current landscape of ODQA benchmarking by reviewing 52 datasets and 20 evaluation techniques across textual and multimodal modalities. We introduce a novel taxonomy for ODQA datasets that incorporates both the modality and difficulty of the question types. Additionally, we present a structured organization of ODQA evaluation metrics along with a critical analysis of their inherent trade-offs. Our study aims to empower researchers by providing a framework for the robust evaluation of modern question-answering systems. We conclude by identifying the current challenges and outlining promising avenues for future research and development. | [
"['Akchay Srivastava' 'Atif Memon']"
] |
null | null | 2406.13246 | null | null | http://arxiv.org/pdf/2406.13246v1 | 2024-06-19T06:15:26Z | 2024-06-19T06:15:26Z | GSR-BENCH: A Benchmark for Grounded Spatial Reasoning Evaluation via
Multimodal LLMs | The ability to understand and reason about spatial relationships between objects in images is an important component of visual reasoning. This skill rests on the ability to recognize and localize objects of interest and determine their spatial relation. Early vision and language models (VLMs) have been shown to struggle to recognize spatial relations. We extend the previously released What'sUp dataset and propose a novel comprehensive evaluation for spatial relationship understanding that highlights the strengths and weaknesses of 27 different models. In addition to the VLMs evaluated in What'sUp, our extensive evaluation encompasses 3 classes of Multimodal LLMs (MLLMs) that vary in their parameter sizes (ranging from 7B to 110B), training/instruction-tuning methods, and visual resolution to benchmark their performances and scrutinize the scaling laws in this task. | [
"['Navid Rajabi' 'Jana Kosecka']"
] |
null | null | 2406.13250 | null | null | http://arxiv.org/pdf/2406.13250v1 | 2024-06-19T06:20:22Z | 2024-06-19T06:20:22Z | LangTopo: Aligning Language Descriptions of Graphs with Tokenized
Topological Modeling | Recently, large language models (LLMs) have been widely researched in the field of graph machine learning due to their outstanding abilities in language comprehension and learning. However, the significant gap between natural language tasks and topological structure modeling poses a nonnegligible challenge. Specifically, since natural language descriptions are not sufficient for LLMs to understand and process graph-structured data, fine-tuned LLMs perform even worse than some traditional GNN models on graph tasks, lacking inherent modeling capabilities for graph structures. Existing research overly emphasizes LLMs' understanding of semantic information captured by external models, while inadequately exploring graph topological structure modeling, thereby overlooking the genuine capabilities that LLMs lack. Consequently, in this paper, we introduce a new framework, LangTopo, which aligns graph structure modeling with natural language understanding at the token level. LangTopo quantifies the graph structure modeling capabilities of GNNs and LLMs by constructing a codebook for the graph modality and performs consistency maximization. This process aligns the text description of LLM with the topological modeling of GNN, allowing LLM to learn the ability of GNN to capture graph structures, enabling LLM to handle graph-structured data independently. We demonstrate the effectiveness of our proposed method on multiple datasets. | [
"['Zhong Guan' 'Hongke Zhao' 'Likang Wu' 'Ming He' 'Jianpin Fan']"
] |
null | null | 2406.13257 | null | null | http://arxiv.org/pdf/2406.13257v1 | 2024-06-19T06:45:19Z | 2024-06-19T06:45:19Z | Reasoning with trees: interpreting CNNs using hierarchies | Challenges persist in providing interpretable explanations for neural network reasoning in explainable AI (xAI). Existing methods like Integrated Gradients produce noisy maps, and LIME, while intuitive, may deviate from the model's reasoning. We introduce a framework that uses hierarchical segmentation techniques for faithful and interpretable explanations of Convolutional Neural Networks (CNNs). Our method constructs model-based hierarchical segmentations that maintain the model's reasoning fidelity and allows both human-centric and model-centric segmentation. This approach offers multiscale explanations, aiding bias identification and enhancing understanding of neural network decision-making. Experiments show that our framework, xAiTrees, delivers highly interpretable and faithful model explanations, not only surpassing traditional xAI methods but shedding new light on a novel approach to enhancing xAI interpretability. Code at: https://github.com/CarolMazini/reasoning_with_trees . | [
"['Caroline Mazini Rodrigues' 'Nicolas Boutry' 'Laurent Najman']"
] |
null | null | 2406.13262 | null | null | http://arxiv.org/abs/2406.13262v2 | 2024-06-24T09:30:24Z | 2024-06-19T06:47:35Z | Machine Learning Applications of Quantum Computing: A Review | At the intersection of quantum computing and machine learning, this review paper explores the transformative impact these technologies are having on the capabilities of data processing and analysis, far surpassing the bounds of traditional computational methods. Drawing upon an in-depth analysis of 32 seminal papers, this review delves into the interplay between quantum computing and machine learning, focusing on transcending the limitations of classical computing in advanced data processing and applications. This review emphasizes the potential of quantum-enhanced methods in enhancing cybersecurity, a critical sector that stands to benefit significantly from these advancements. The literature review, primarily leveraging Science Direct as an academic database, delves into the transformative effects of quantum technologies on machine learning, drawing insights from a diverse collection of studies and scholarly articles. While the focus is primarily on the growing significance of quantum computing in cybersecurity, the review also acknowledges the promising implications for other sectors as the field matures. Our systematic approach categorizes sources based on quantum machine learning algorithms, applications, challenges, and potential future developments, uncovering that quantum computing is increasingly being implemented in practical machine learning scenarios. The review highlights advancements in quantum-enhanced machine learning algorithms and their potential applications in sectors such as cybersecurity, emphasizing the need for industry-specific solutions while considering ethical and security concerns. By presenting an overview of the current state and projecting future directions, the paper sets a foundation for ongoing research and strategic advancement in quantum machine learning. | [
"['Thien Nguyen' 'Tuomo Sipola' 'Jari Hautamäki']"
] |
null | null | 2406.13264 | null | null | http://arxiv.org/pdf/2406.13264v1 | 2024-06-19T06:50:15Z | 2024-06-19T06:50:15Z | Do Multimodal Foundation Models Understand Enterprise Workflows? A
Benchmark for Business Process Management Tasks | Existing ML benchmarks lack the depth and diversity of annotations needed for evaluating models on business process management (BPM) tasks. BPM is the practice of documenting, measuring, improving, and automating enterprise workflows. However, research has focused almost exclusively on one task - full end-to-end automation using agents based on multimodal foundation models (FMs) like GPT-4. This focus on automation ignores the reality of how most BPM tools are applied today - simply documenting the relevant workflow takes 60% of the time of the typical process optimization project. To address this gap we present WONDERBREAD, the first benchmark for evaluating multimodal FMs on BPM tasks beyond automation. Our contributions are: (1) a dataset containing 2928 documented workflow demonstrations; (2) 6 novel BPM tasks sourced from real-world applications ranging from workflow documentation to knowledge transfer to process improvement; and (3) an automated evaluation harness. Our benchmark shows that while state-of-the-art FMs can automatically generate documentation (e.g. recalling 88% of the steps taken in a video demonstration of a workflow), they struggle to re-apply that knowledge towards finer-grained validation of workflow completion (F1 < 0.3). We hope WONDERBREAD encourages the development of more "human-centered" AI tooling for enterprise applications and furthers the exploration of multimodal FMs for the broader universe of BPM tasks. We publish our dataset and experiments here: https://github.com/HazyResearch/wonderbread | [
"['Michael Wornow' 'Avanika Narayan' 'Ben Viggiano' 'Ishan S. Khare'\n 'Tathagat Verma' 'Tibor Thompson' 'Miguel Angel Fuentes Hernandez'\n 'Sudharsan Sundar' 'Chloe Trujillo' 'Krrish Chawla' 'Rongfei Lu'\n 'Justin Shen' 'Divya Nagaraj' 'Joshua Martinez' 'Vardhan Agrawal'\n 'Althea Hudson' 'Nigam H. Shah' 'Christopher Re']"
] |
null | null | 2406.13265 | null | null | http://arxiv.org/pdf/2406.13265v1 | 2024-06-19T06:53:09Z | 2024-06-19T06:53:09Z | Molecule Graph Networks with Many-body Equivariant Interactions | Message passing neural networks have demonstrated significant efficacy in predicting molecular interactions. Introducing equivariant vectorial representations augments expressivity by capturing geometric data symmetries, thereby improving model accuracy. However, two-body bond vectors in opposition may cancel each other out during message passing, leading to the loss of directional information on their shared node. In this study, we develop Equivariant N-body Interaction Networks (ENINet) that explicitly integrates equivariant many-body interactions to preserve directional information in the message passing scheme. Experiments indicate that integrating many-body equivariant representations enhances prediction accuracy across diverse scalar and tensorial quantum chemical properties. Ablation studies show an average performance improvement of 7.9% across 11 out of 12 properties in QM9, 27.9% in forces in MD17, and 11.3% in polarizabilities (CCSD) in QM7b. | [
"['Zetian Mao' 'Jiawen Li' 'Chen Liang' 'Diptesh Das' 'Masato Sumita'\n 'Koji Tsuda']"
] |
null | null | 2406.13283 | null | null | http://arxiv.org/pdf/2406.13283v2 | 2024-07-11T17:10:24Z | 2024-06-19T07:23:51Z | Large-Scale Dataset Pruning in Adversarial Training through Data
Importance Extrapolation | Their vulnerability to small, imperceptible attacks limits the adoption of deep learning models to real-world systems. Adversarial training has proven to be one of the most promising strategies against these attacks, at the expense of a substantial increase in training time. With the ongoing trend of integrating large-scale synthetic data this is only expected to increase even further. Thus, the need for data-centric approaches that reduce the number of training samples while maintaining accuracy and robustness arises. While data pruning and active learning are prominent research topics in deep learning, they are as of now largely unexplored in the adversarial training literature. We address this gap and propose a new data pruning strategy based on extrapolating data importance scores from a small set of data to a larger set. In an empirical evaluation, we demonstrate that extrapolation-based pruning can efficiently reduce dataset size while maintaining robustness. | [
"['Björn Nieth' 'Thomas Altstidl' 'Leo Schwinn' 'Björn Eskofier']"
] |
null | null | 2406.13294 | null | null | http://arxiv.org/pdf/2406.13294v1 | 2024-06-19T07:32:55Z | 2024-06-19T07:32:55Z | Enhancing Cross-Prompt Transferability in Vision-Language Models through
Contextual Injection of Target Tokens | Vision-language models (VLMs) seamlessly integrate visual and textual data to perform tasks such as image classification, caption generation, and visual question answering. However, adversarial images often struggle to deceive all prompts effectively in the context of cross-prompt migration attacks, as the probability distribution of the tokens in these images tends to favor the semantics of the original image rather than the target tokens. To address this challenge, we propose a Contextual-Injection Attack (CIA) that employs gradient-based perturbation to inject target tokens into both visual and textual contexts, thereby improving the probability distribution of the target tokens. By shifting the contextual semantics towards the target tokens instead of the original image semantics, CIA enhances the cross-prompt transferability of adversarial images.Extensive experiments on the BLIP2, InstructBLIP, and LLaVA models show that CIA outperforms existing methods in cross-prompt transferability, demonstrating its potential for more effective adversarial strategies in VLMs. | [
"['Xikang Yang' 'Xuehai Tang' 'Fuqing Zhu' 'Jizhong Han' 'Songlin Hu']"
] |
null | null | 2406.13300 | null | null | http://arxiv.org/pdf/2406.13300v1 | 2024-06-19T07:40:37Z | 2024-06-19T07:40:37Z | LightGBM robust optimization algorithm based on topological data
analysis | To enhance the robustness of the Light Gradient Boosting Machine (LightGBM) algorithm for image classification, a topological data analysis (TDA)-based robustness optimization algorithm for LightGBM, TDA-LightGBM, is proposed to address the interference of noise on image classification. Initially, the method partitions the feature engineering process into two streams: pixel feature stream and topological feature stream for feature extraction respectively. Subsequently, these pixel and topological features are amalgamated into a comprehensive feature vector, serving as the input for LightGBM in image classification tasks. This fusion of features not only encompasses traditional feature engineering methodologies but also harnesses topological structure information to more accurately encapsulate the intrinsic features of the image. The objective is to surmount challenges related to unstable feature extraction and diminished classification accuracy induced by data noise in conventional image processing. Experimental findings substantiate that TDA-LightGBM achieves a 3% accuracy improvement over LightGBM on the SOCOFing dataset across five classification tasks under noisy conditions. In noise-free scenarios, TDA-LightGBM exhibits a 0.5% accuracy enhancement over LightGBM on two classification tasks, achieving a remarkable accuracy of 99.8%. Furthermore, the method elevates the classification accuracy of the Ultrasound Breast Images for Breast Cancer dataset and the Masked CASIA WebFace dataset by 6% and 15%, respectively, surpassing LightGBM in the presence of noise. These empirical results underscore the efficacy of the TDA-LightGBM approach in fortifying the robustness of LightGBM by integrating topological features, thereby augmenting the performance of image classification tasks amidst data perturbations. | [
"['Han Yang' 'Guangjun Qin' 'Ziyuan Liu' 'Yongqing Hu' 'Qinglong Dai']"
] |
null | null | 2406.13329 | null | null | http://arxiv.org/pdf/2406.13329v1 | 2024-06-19T08:22:51Z | 2024-06-19T08:22:51Z | On rough mereology and VC-dimension in treatment of decision prediction
for open world decision systems | Given a raw knowledge in the form of a data table/a decision system, one is facing two possible venues. One, to treat the system as closed, i.e., its universe does not admit new objects, or, to the contrary, its universe is open on admittance of new objects. In particular, one may obtain new objects whose sets of values of features are new to the system. In this case the problem is to assign a decision value to any such new object. This problem is somehow resolved in the rough set theory, e.g., on the basis of similarity of the value set of a new object to value sets of objects already assigned a decision value. It is crucial for online learning when each new object must have a predicted decision value. There is a vast literature on various methods for decision prediction for new yet unseen object. The approach we propose is founded in the theory of rough mereology and it requires a theory of sets/concepts, and, we root our theory in classical set theory of Syllogistic within which we recall the theory of parts known as Mereology. Then, we recall our theory of Rough Mereology along with the theory of weight assignment to the Tarski algebra of Mereology. This allows us to introduce the notion of a part to a degree. Once we have defined basics of Mereology and rough Mereology, we recall our theory of weight assignment to elements of the Boolean algebra within Mereology and this allows us to define the relation of parts to the degree and we apply this notion in a procedure to select a decision for new yet unseen objects. In selecting a plausible candidate which would pass its decision value to the new object, we employ the notion of Vapnik - Chervonenkis dimension in order to select at the first stage the candidate with the largest VC-dimension of the family of its $varepsilon$-components for some choice of $varepsilon$. | [
"['Lech T. Polkowski']"
] |
null | null | 2406.13337 | null | null | http://arxiv.org/pdf/2406.13337v1 | 2024-06-19T08:39:09Z | 2024-06-19T08:39:09Z | Medical Spoken Named Entity Recognition | Spoken Named Entity Recognition (NER) aims to extracting named entities from speech and categorizing them into types like person, location, organization, etc. In this work, we present VietMed-NER - the first spoken NER dataset in the medical domain. To our best knowledge, our real-world dataset is the largest spoken NER dataset in the world in terms of the number of entity types, featuring 18 distinct types. Secondly, we present baseline results using various state-of-the-art pre-trained models: encoder-only and sequence-to-sequence. We found that pre-trained multilingual models XLM-R outperformed all monolingual models on both reference text and ASR output. Also in general, encoders perform better than sequence-to-sequence models for the NER task. By simply translating, the transcript is applicable not just to Vietnamese but to other languages as well. All code, data and models are made publicly available here: https://github.com/leduckhai/MultiMed | [
"['Khai Le-Duc']"
] |
null | null | 2406.13348 | null | null | http://arxiv.org/pdf/2406.13348v1 | 2024-06-19T08:51:54Z | 2024-06-19T08:51:54Z | Textual Unlearning Gives a False Sense of Unlearning | Language models (LMs) are susceptible to "memorizing" training data, including a large amount of private or copyright-protected content. To safeguard the right to be forgotten (RTBF), machine unlearning has emerged as a promising method for LMs to efficiently "forget" sensitive training content and mitigate knowledge leakage risks. However, despite its good intentions, could the unlearning mechanism be counterproductive? In this paper, we propose the Textual Unlearning Leakage Attack (TULA), where an adversary can infer information about the unlearned data only by accessing the models before and after unlearning. Furthermore, we present variants of TULA in both black-box and white-box scenarios. Through various experimental results, we critically demonstrate that machine unlearning amplifies the risk of knowledge leakage from LMs. Specifically, TULA can increase an adversary's ability to infer membership information about the unlearned data by more than 20% in black-box scenario. Moreover, TULA can even reconstruct the unlearned data directly with more than 60% accuracy with white-box access. Our work is the first to reveal that machine unlearning in LMs can inversely create greater knowledge risks and inspire the development of more secure unlearning mechanisms. | [
"['Jiacheng Du' 'Zhibo Wang' 'Kui Ren']"
] |
null | null | 2406.13351 | null | null | http://arxiv.org/pdf/2406.13351v1 | 2024-06-19T08:55:40Z | 2024-06-19T08:55:40Z | A Resource-Adaptive Approach for Federated Learning under
Resource-Constrained Environments | The paper studies a fundamental federated learning (FL) problem involving multiple clients with heterogeneous constrained resources. Compared with the numerous training parameters, the computing and communication resources of clients are insufficient for fast local training and real-time knowledge sharing. Besides, training on clients with heterogeneous resources may result in the straggler problem. To address these issues, we propose Fed-RAA: a Resource-Adaptive Asynchronous Federated learning algorithm. Different from vanilla FL methods, where all parameters are trained by each participating client regardless of resource diversity, Fed-RAA adaptively allocates fragments of the global model to clients based on their computing and communication capabilities. Each client then individually trains its assigned model fragment and asynchronously uploads the updated result. Theoretical analysis confirms the convergence of our approach. Additionally, we design an online greedy-based algorithm for fragment allocation in Fed-RAA, achieving fairness comparable to an offline strategy. We present numerical results on MNIST, CIFAR-10, and CIFAR-100, along with necessary comparisons and ablation studies, demonstrating the advantages of our work. To the best of our knowledge, this paper represents the first resource-adaptive asynchronous method for fragment-based FL with guaranteed theoretical convergence. | [
"['Ruirui Zhang' 'Xingze Wu' 'Yifei Zou' 'Zhenzhen Xie' 'Peng Li'\n 'Xiuzhen Cheng' 'Dongxiao Yu']"
] |
null | null | 2406.13352 | null | null | http://arxiv.org/pdf/2406.13352v1 | 2024-06-19T08:55:56Z | 2024-06-19T08:55:56Z | AgentDojo: A Dynamic Environment to Evaluate Attacks and Defenses for
LLM Agents | AI agents aim to solve complex tasks by combining text-based reasoning with external tool calls. Unfortunately, AI agents are vulnerable to prompt injection attacks where data returned by external tools hijacks the agent to execute malicious tasks. To measure the adversarial robustness of AI agents, we introduce AgentDojo, an evaluation framework for agents that execute tools over untrusted data. To capture the evolving nature of attacks and defenses, AgentDojo is not a static test suite, but rather an extensible environment for designing and evaluating new agent tasks, defenses, and adaptive attacks. We populate the environment with 97 realistic tasks (e.g., managing an email client, navigating an e-banking website, or making travel bookings), 629 security test cases, and various attack and defense paradigms from the literature. We find that AgentDojo poses a challenge for both attacks and defenses: state-of-the-art LLMs fail at many tasks (even in the absence of attacks), and existing prompt injection attacks break some security properties but not all. We hope that AgentDojo can foster research on new design principles for AI agents that solve common tasks in a reliable and robust manner. We release the code for AgentDojo at https://github.com/ethz-spylab/agentdojo. | [
"['Edoardo Debenedetti' 'Jie Zhang' 'Mislav Balunović'\n 'Luca Beurer-Kellner' 'Marc Fischer' 'Florian Tramèr']"
] |
null | null | 2406.13356 | null | null | http://arxiv.org/pdf/2406.13356v1 | 2024-06-19T09:03:21Z | 2024-06-19T09:03:21Z | Jogging the Memory of Unlearned Model Through Targeted Relearning Attack | Machine unlearning is a promising approach to mitigate undesirable memorization of training data in ML models. However, in this work we show that existing approaches for unlearning in LLMs are surprisingly susceptible to a simple set of targeted relearning attacks. With access to only a small and potentially loosely related set of data, we find that we can 'jog' the memory of unlearned models to reverse the effects of unlearning. We formalize this unlearning-relearning pipeline, explore the attack across three popular unlearning benchmarks, and discuss future directions and guidelines that result from our study. | [
"['Shengyuan Hu' 'Yiwei Fu' 'Zhiwei Steven Wu' 'Virginia Smith']"
] |
null | null | 2406.13361 | null | null | http://arxiv.org/pdf/2406.13361v1 | 2024-06-19T09:06:24Z | 2024-06-19T09:06:24Z | Improving Zero-Shot Cross-Lingual Transfer via Progressive
Code-Switching | Code-switching is a data augmentation scheme mixing words from multiple languages into source lingual text. It has achieved considerable generalization performance of cross-lingual transfer tasks by aligning cross-lingual contextual word representations. However, uncontrolled and over-replaced code-switching would augment dirty samples to model training. In other words, the excessive code-switching text samples will negatively hurt the models' cross-lingual transferability. To this end, we propose a Progressive Code-Switching (PCS) method to gradually generate moderately difficult code-switching examples for the model to discriminate from easy to hard. The idea is to incorporate progressively the preceding learned multilingual knowledge using easier code-switching data to guide model optimization on succeeding harder code-switching data. Specifically, we first design a difficulty measurer to measure the impact of replacing each word in a sentence based on the word relevance score. Then a code-switcher generates the code-switching data of increasing difficulty via a controllable temperature variable. In addition, a training scheduler decides when to sample harder code-switching data for model training. Experiments show our model achieves state-of-the-art results on three different zero-shot cross-lingual transfer tasks across ten languages. | [
"['Zhuoran Li' 'Chunming Hu' 'Junfan Chen' 'Zhijun Chen' 'Xiaohui Guo'\n 'Richong Zhang']"
] |
null | null | 2406.13362 | null | null | http://arxiv.org/pdf/2406.13362v1 | 2024-06-19T09:07:31Z | 2024-06-19T09:07:31Z | VisualRWKV: Exploring Recurrent Neural Networks for Visual Language
Models | Visual Language Models (VLMs) have rapidly progressed with the recent success of large language models. However, there have been few attempts to incorporate efficient linear Recurrent Neural Networks (RNNs) architectures into VLMs. In this study, we introduce VisualRWKV, the first application of a linear RNN model to multimodal learning tasks, leveraging the pre-trained RWKV language model. We propose a data-dependent recurrence and sandwich prompts to enhance our modeling capabilities, along with a 2D image scanning mechanism to enrich the processing of visual sequences. Extensive experiments demonstrate that VisualRWKV achieves competitive performance compared to Transformer-based models like LLaVA-1.5 on various benchmarks. To facilitate further research and analysis, we have made the checkpoints and the associated code publicly accessible at the following GitHub repository: href{https://github.com/howard-hou/VisualRWKV}{https://github.com/howard-hou/VisualRWKV}. | [
"['Haowen Hou' 'Peigen Zeng' 'Fei Ma' 'Fei Richard Yu']"
] |
null | null | 2406.13365 | null | null | http://arxiv.org/pdf/2406.13365v1 | 2024-06-19T09:09:46Z | 2024-06-19T09:09:46Z | PPT-GNN: A Practical Pre-Trained Spatio-Temporal Graph Neural Network
for Network Security | Recent works have demonstrated the potential of Graph Neural Networks (GNN) for network intrusion detection. Despite their advantages, a significant gap persists between real-world scenarios, where detection speed is critical, and existing proposals, which operate on large graphs representing several hours of traffic. This gap results in unrealistic operational conditions and impractical detection delays. Moreover, existing models do not generalize well across different networks, hampering their deployment in production environments. To address these issues, we introduce PPTGNN, a practical spatio-temporal GNN for intrusion detection. PPTGNN enables near real-time predictions, while better capturing the spatio-temporal dynamics of network attacks. PPTGNN employs self-supervised pre-training for improved performance and reduced dependency on labeled data. We evaluate PPTGNN on three public datasets and show that it significantly outperforms state-of-the-art models, such as E-ResGAT and E-GraphSAGE, with an average accuracy improvement of 10.38%. Finally, we show that a pre-trained PPTGNN can easily be fine-tuned to unseen networks with minimal labeled examples. This highlights the potential of PPTGNN as a general, large-scale pre-trained model that can effectively operate in diverse network environments. | [
"['Louis Van Langendonck' 'Ismael Castell-Uroz' 'Pere Barlet-Ros']"
] |
null | null | 2406.13369 | null | null | http://arxiv.org/pdf/2406.13369v1 | 2024-06-19T09:11:03Z | 2024-06-19T09:11:03Z | Effective Edge-wise Representation Learning in Edge-Attributed Bipartite
Graphs | Graph representation learning (GRL) is to encode graph elements into informative vector representations, which can be used in downstream tasks for analyzing graph-structured data and has seen extensive applications in various domains. However, the majority of extant studies on GRL are geared towards generating node representations, which cannot be readily employed to perform edge-based analytics tasks in edge-attributed bipartite graphs (EABGs) that pervade the real world, e.g., spam review detection in customer-product reviews and identifying fraudulent transactions in user-merchant networks. Compared to node-wise GRL, learning edge representations (ERL) on such graphs is challenging due to the need to incorporate the structure and attribute semantics from the perspective of edges while considering the separate influence of two heterogeneous node sets U and V in bipartite graphs. To our knowledge, despite its importance, limited research has been devoted to this frontier, and existing workarounds all suffer from sub-par results. Motivated by this, this paper designs EAGLE, an effective ERL method for EABGs. Building on an in-depth and rigorous theoretical analysis, we propose the factorized feature propagation (FFP) scheme for edge representations with adequate incorporation of long-range dependencies of edges/features without incurring tremendous computation overheads. We further ameliorate FFP as a dual-view FFP by taking into account the influences from nodes in U and V severally in ERL. Extensive experiments on 5 real datasets showcase the effectiveness of the proposed EAGLE models in semi-supervised edge classification tasks. In particular, EAGLE can attain a considerable gain of at most 38.11% in AP and 1.86% in AUC when compared to the best baselines. | [
"['Hewen Wang' 'Renchi Yang' 'Xiaokui Xiao']"
] |
null | null | 2406.13371 | null | null | http://arxiv.org/abs/2406.13371v1 | 2024-06-19T09:14:40Z | 2024-06-19T09:14:40Z | Identifiable Causal Representation Learning: Unsupervised, Multi-View,
and Multi-Environment | Causal models provide rich descriptions of complex systems as sets of mechanisms by which each variable is influenced by its direct causes. They support reasoning about manipulating parts of the system and thus hold promise for addressing some of the open challenges of artificial intelligence (AI), such as planning, transferring knowledge in changing environments, or robustness to distribution shifts. However, a key obstacle to more widespread use of causal models in AI is the requirement that the relevant variables be specified a priori, which is typically not the case for the high-dimensional, unstructured data processed by modern AI systems. At the same time, machine learning (ML) has proven quite successful at automatically extracting useful and compact representations of such complex data. Causal representation learning (CRL) aims to combine the core strengths of ML and causality by learning representations in the form of latent variables endowed with causal model semantics. In this thesis, we study and present new results for different CRL settings. A central theme is the question of identifiability: Given infinite data, when are representations satisfying the same learning objective guaranteed to be equivalent? This is an important prerequisite for CRL, as it formally characterises if and when a learning task is, at least in principle, feasible. Since learning causal models, even without a representation learning component, is notoriously difficult, we require additional assumptions on the model class or rich data beyond the classical i.i.d. setting. By partially characterising identifiability for different settings, this thesis investigates what is possible for CRL without direct supervision, and thus contributes to its theoretical foundations. Ideally, the developed insights can help inform data collection practices or inspire the design of new practical estimation methods. | [
"['Julius von Kügelgen']"
] |
null | null | 2406.13376 | null | null | http://arxiv.org/pdf/2406.13376v1 | 2024-06-19T09:16:38Z | 2024-06-19T09:16:38Z | Efficient Offline Reinforcement Learning: The Critic is Critical | Recent work has demonstrated both benefits and limitations from using supervised approaches (without temporal-difference learning) for offline reinforcement learning. While off-policy reinforcement learning provides a promising approach for improving performance beyond supervised approaches, we observe that training is often inefficient and unstable due to temporal difference bootstrapping. In this paper we propose a best-of-both approach by first learning the behavior policy and critic with supervised learning, before improving with off-policy reinforcement learning. Specifically, we demonstrate improved efficiency by pre-training with a supervised Monte-Carlo value-error, making use of commonly neglected downstream information from the provided offline trajectories. We find that we are able to more than halve the training time of the considered offline algorithms on standard benchmarks, and surprisingly also achieve greater stability. We further build on the importance of having consistent policy and value functions to propose novel hybrid algorithms, TD3+BC+CQL and EDAC+BC, that regularize both the actor and the critic towards the behavior policy. This helps to more reliably improve on the behavior policy when learning from limited human demonstrations. Code is available at https://github.com/AdamJelley/EfficientOfflineRL | [
"['Adam Jelley' 'Trevor McInroe' 'Sam Devlin' 'Amos Storkey']"
] |
null | null | 2406.13389 | null | null | http://arxiv.org/abs/2406.13389v1 | 2024-06-19T09:30:11Z | 2024-06-19T09:30:11Z | Unifying Mixed Gas Adsorption in Molecular Sieve Membranes and MOFs
using Machine Learning | Recent machine learning models to accurately obtain gas adsorption isotherms focus on polymers or metal-organic frameworks (MOFs) separately. The difficulty in creating a unified model that can predict the adsorption trends in both types of adsorbents is challenging, owing to the diversity in their chemical structures. Moreover, models trained only on single gas adsorption data are incapable of predicting adsorption isotherms for binary gas mixtures. In this work, we address these problems using feature vectors comprising only the physical properties of the gas mixtures and adsorbents. Our model is trained on adsorption isotherms of both single and binary mixed gases inside carbon molecular sieving membrane (CMSM), together with data available from CoRE MOF database. The trained models are capable of accurately predicting the adsorption trends in both classes of materials, for both pure and binary components. ML architecture designed for one class of material, is not suitable for predicting the other class, even after proper training, signifying that the model must be trained jointly for proper predictions and transferability. The model is used to predict with good accuracy the CO2 uptake inside CALF-20 framework. This work opens up a new avenue for predicting complex adsorption processes for gas mixtures in a wide range of materials. | [
"['Subhadeep Dasgupta' 'Amal R S' 'Prabal K. Maiti']"
] |
null | null | 2406.13411 | null | null | http://arxiv.org/pdf/2406.13411v2 | 2024-06-21T04:11:33Z | 2024-06-19T10:02:54Z | Composite Concept Extraction through Backdooring | Learning composite concepts, such as textquotedbl red cartextquotedbl , from individual examples -- like a white car representing the concept of textquotedbl cartextquotedbl{} and a red strawberry representing the concept of textquotedbl redtextquotedbl -- is inherently challenging. This paper introduces a novel method called Composite Concept Extractor (CoCE), which leverages techniques from traditional backdoor attacks to learn these composite concepts in a zero-shot setting, requiring only examples of individual concepts. By repurposing the trigger-based model backdooring mechanism, we create a strategic distortion in the manifold of the target object (e.g., textquotedbl cartextquotedbl ) induced by example objects with the target property (e.g., textquotedbl redtextquotedbl ) from objects textquotedbl red strawberrytextquotedbl , ensuring the distortion selectively affects the target objects with the target property. Contrastive learning is then employed to further refine this distortion, and a method is formulated for detecting objects that are influenced by the distortion. Extensive experiments with in-depth analysis across different datasets demonstrate the utility and applicability of our proposed approach. | [
"['Banibrata Ghosh' 'Haripriya Harikumar' 'Khoa D Doan' 'Svetha Venkatesh'\n 'Santu Rana']"
] |
null | null | 2406.13415 | null | null | http://arxiv.org/pdf/2406.13415v1 | 2024-06-19T10:11:37Z | 2024-06-19T10:11:37Z | Factual Confidence of LLMs: on Reliability and Robustness of Current
Estimators | Large Language Models (LLMs) tend to be unreliable in the factuality of their answers. To address this problem, NLP researchers have proposed a range of techniques to estimate LLM's confidence over facts. However, due to the lack of a systematic comparison, it is not clear how the different methods compare to one another. To fill this gap, we present a survey and empirical comparison of estimators of factual confidence. We define an experimental framework allowing for fair comparison, covering both fact-verification and question answering. Our experiments across a series of LLMs indicate that trained hidden-state probes provide the most reliable confidence estimates, albeit at the expense of requiring access to weights and training data. We also conduct a deeper assessment of factual confidence by measuring the consistency of model behavior under meaning-preserving variations in the input. We find that the confidence of LLMs is often unstable across semantically equivalent inputs, suggesting that there is much room for improvement of the stability of models' parametric knowledge. Our code is available at (https://github.com/amazon-science/factual-confidence-of-llms). | [
"['Matéo Mahaut' 'Laura Aina' 'Paula Czarnowska' 'Momchil Hardalov'\n 'Thomas Müller' 'Lluís Màrquez']"
] |
null | null | 2406.13424 | null | null | http://arxiv.org/pdf/2406.13424v1 | 2024-06-19T10:30:56Z | 2024-06-19T10:30:56Z | Towards a multimodal framework for remote sensing image change retrieval
and captioning | Recently, there has been increasing interest in multimodal applications that integrate text with other modalities, such as images, audio and video, to facilitate natural language interactions with multimodal AI systems. While applications involving standard modalities have been extensively explored, there is still a lack of investigation into specific data modalities such as remote sensing (RS) data. Despite the numerous potential applications of RS data, including environmental protection, disaster monitoring and land planning, available solutions are predominantly focused on specific tasks like classification, captioning and retrieval. These solutions often overlook the unique characteristics of RS data, such as its capability to systematically provide information on the same geographical areas over time. This ability enables continuous monitoring of changes in the underlying landscape. To address this gap, we propose a novel foundation model for bi-temporal RS image pairs, in the context of change detection analysis, leveraging Contrastive Learning and the LEVIR-CC dataset for both captioning and text-image retrieval. By jointly training a contrastive encoder and captioning decoder, our model add text-image retrieval capabilities, in the context of bi-temporal change detection, while maintaining captioning performances that are comparable to the state of the art. We release the source code and pretrained weights at: https://github.com/rogerferrod/RSICRC. | [
"['Roger Ferrod' 'Luigi Di Caro' 'Dino Ienco']"
] |
null | null | 2406.13425 | null | null | http://arxiv.org/pdf/2406.13425v1 | 2024-06-19T10:31:57Z | 2024-06-19T10:31:57Z | Coupled Input-Output Dimension Reduction: Application to Goal-oriented
Bayesian Experimental Design and Global Sensitivity Analysis | We introduce a new method to jointly reduce the dimension of the input and output space of a high-dimensional function. Choosing a reduced input subspace influences which output subspace is relevant and vice versa. Conventional methods focus on reducing either the input or output space, even though both are often reduced simultaneously in practice. Our coupled approach naturally supports goal-oriented dimension reduction, where either an input or output quantity of interest is prescribed. We consider, in particular, goal-oriented sensor placement and goal-oriented sensitivity analysis, which can be viewed as dimension reduction where the most important output or, respectively, input components are chosen. Both applications present difficult combinatorial optimization problems with expensive objectives such as the expected information gain and Sobol indices. By optimizing gradient-based bounds, we can determine the most informative sensors and most sensitive parameters as the largest diagonal entries of some diagnostic matrices, thus bypassing the combinatorial optimization and objective evaluation. | [
"['Qiao Chen' 'Elise Arnaud' 'Ricardo Baptista' 'Olivier Zahm']"
] |
null | null | 2406.13427 | null | null | http://arxiv.org/pdf/2406.13427v1 | 2024-06-19T10:36:38Z | 2024-06-19T10:36:38Z | Are Logistic Models Really Interpretable? | The demand for open and trustworthy AI models points towards widespread publishing of model weights. Consumers of these model weights must be able to act accordingly with the information provided. That said, one of the simplest AI classification models, Logistic Regression (LR), has an unwieldy interpretation of its model weights, with greater difficulties when extending LR to generalised additive models. In this work, we show via a User Study that skilled participants are unable to reliably reproduce the action of small LR models given the trained parameters. As an antidote to this, we define Linearised Additive Models (LAMs), an optimal piecewise linear approximation that augments any trained additive model equipped with a sigmoid link function, requiring no retraining. We argue that LAMs are more interpretable than logistic models -- survey participants are shown to solve model reasoning tasks with LAMs much more accurately than with LR given the same information. Furthermore, we show that LAMs do not suffer from large performance penalties in terms of ROC-AUC and calibration with respect to their logistic counterparts on a broad suite of public financial modelling data. | [
"['Danial Dervovic' 'Freddy Lécué' 'Nicolás Marchesotti'\n 'Daniele Magazzeni']"
] |
null | null | 2406.13433 | null | null | http://arxiv.org/pdf/2406.13433v1 | 2024-06-19T10:47:00Z | 2024-06-19T10:47:00Z | Certificates of Differential Privacy and Unlearning for Gradient-Based
Training | Proper data stewardship requires that model owners protect the privacy of individuals' data used during training. Whether through anonymization with differential privacy or the use of unlearning in non-anonymized settings, the gold-standard techniques for providing privacy guarantees can come with significant performance penalties or be too weak to provide practical assurances. In part, this is due to the fact that the guarantee provided by differential privacy represents the worst-case privacy leakage for any individual, while the true privacy leakage of releasing the prediction for a given individual might be substantially smaller or even, as we show, non-existent. This work provides a novel framework based on convex relaxations and bounds propagation that can compute formal guarantees (certificates) that releasing specific predictions satisfies $epsilon=0$ privacy guarantees or do not depend on data that is subject to an unlearning request. Our framework offers a new verification-centric approach to privacy and unlearning guarantees, that can be used to further engender user trust with tighter privacy guarantees, provide formal proofs of robustness to certain membership inference attacks, identify potentially vulnerable records, and enhance current unlearning approaches. We validate the effectiveness of our approach on tasks from financial services, medical imaging, and natural language processing. | [
"['Matthew Wicker' 'Philip Sosnin' 'Adrianna Janik' 'Mark N. Müller'\n 'Adrian Weller' 'Calvin Tsay']"
] |
null | null | 2406.13447 | null | null | http://arxiv.org/pdf/2406.13447v2 | 2024-07-04T15:08:50Z | 2024-06-19T11:15:01Z | High-probability minimax lower bounds | The minimax risk is often considered as a gold standard against which we can compare specific statistical procedures. Nevertheless, as has been observed recently in robust and heavy-tailed estimation problems, the inherent reduction of the (random) loss to its expectation may entail a significant loss of information regarding its tail behaviour. In an attempt to avoid such a loss, we introduce the notion of a minimax quantile, and seek to articulate its dependence on the quantile level. To this end, we develop high-probability variants of the classical Le Cam and Fano methods, as well as a technique to convert local minimax risk lower bounds to lower bounds on minimax quantiles. To illustrate the power of our framework, we deploy our techniques on several examples, recovering recent results in robust mean estimation and stochastic convex optimisation, as well as obtaining several new results in covariance matrix estimation, sparse linear regression, nonparametric density estimation and isotonic regression. Our overall goal is to argue that minimax quantiles can provide a finer-grained understanding of the difficulty of statistical problems, and that, in wide generality, lower bounds on these quantities can be obtained via user-friendly tools. | [
"['Tianyi Ma' 'Kabir A. Verchand' 'Richard J. Samworth']"
] |
null | null | 2406.13469 | null | null | http://arxiv.org/pdf/2406.13469v1 | 2024-06-19T11:50:09Z | 2024-06-19T11:50:09Z | Encoder vs Decoder: Comparative Analysis of Encoder and Decoder Language
Models on Multilingual NLU Tasks | This paper explores the performance of encoder and decoder language models on multilingual Natural Language Understanding (NLU) tasks, with a broad focus on Germanic languages. Building upon the ScandEval benchmark, which initially was restricted to evaluating encoder models, we extend the evaluation framework to include decoder models. We introduce a method for evaluating decoder models on NLU tasks and apply it to the languages Danish, Swedish, Norwegian, Icelandic, Faroese, German, Dutch, and English. Through a series of experiments and analyses, we address key research questions regarding the comparative performance of encoder and decoder models, the impact of NLU task types, and the variation across language resources. Our findings reveal that decoder models can achieve significantly better NLU performance than encoder models, with nuances observed across different tasks and languages. Additionally, we investigate the correlation between decoders and task performance via a UMAP analysis, shedding light on the unique capabilities of decoder and encoder models. This study contributes to a deeper understanding of language model paradigms in NLU tasks and provides valuable insights for model selection and evaluation in multilingual settings. | [
"['Dan Saattrup Nielsen' 'Kenneth Enevoldsen' 'Peter Schneider-Kamp']"
] |
null | null | 2406.13474 | null | null | http://arxiv.org/pdf/2406.13474v1 | 2024-06-19T11:53:21Z | 2024-06-19T11:53:21Z | Attention-aware Post-training Quantization without Backpropagation | Quantization is a promising solution for deploying large-scale language models (LLMs) on resource-constrained devices. Existing quantization approaches, however, rely on gradient-based optimization, regardless of it being post-training quantization (PTQ) or quantization-aware training (QAT), which becomes problematic for hyper-scale LLMs with billions of parameters. This overhead can be alleviated via recently proposed backpropagation-free PTQ methods; however, their performance is somewhat limited by their lack of consideration of inter-layer dependencies. In this paper, we thus propose a novel PTQ algorithm that considers inter-layer dependencies without relying on backpropagation. The fundamental concept involved is the development of attention-aware Hessian matrices, which facilitates the consideration of inter-layer dependencies within the attention module. Extensive experiments demonstrate that the proposed algorithm significantly outperforms conventional PTQ methods, particularly for low bit-widths. | [
"['Junhan Kim' 'Ho-young Kim' 'Eulrang Cho' 'Chungman Lee' 'Joonyoung Kim'\n 'Yongkweon Jeon']"
] |
null | null | 2406.13486 | null | null | http://arxiv.org/pdf/2406.13486v1 | 2024-06-19T12:11:42Z | 2024-06-19T12:11:42Z | Mean-Variance Portfolio Selection in Long-Term Investments with Unknown
Distribution: Online Estimation, Risk Aversion under Ambiguity, and
Universality of Algorithms | The standard approach for constructing a Mean-Variance portfolio involves estimating parameters for the model using collected samples. However, since the distribution of future data may not resemble that of the training set, the out-of-sample performance of the estimated portfolio is worse than one derived with true parameters, which has prompted several innovations for better estimation. Instead of treating the data without a timing aspect as in the common training-backtest approach, this paper adopts a perspective where data gradually and continuously reveal over time. The original model is recast into an online learning framework, which is free from any statistical assumptions, to propose a dynamic strategy of sequential portfolios such that its empirical utility, Sharpe ratio, and growth rate asymptotically achieve those of the true portfolio, derived with perfect knowledge of the future data. When the distribution of future data has a normal shape, the growth rate of wealth is shown to increase by lifting the portfolio along the efficient frontier through the calibration of risk aversion. Since risk aversion cannot be appropriately predetermined, another proposed algorithm updating this coefficient over time forms a dynamic strategy approaching the optimal empirical Sharpe ratio or growth rate associated with the true coefficient. The performance of these proposed strategies is universally guaranteed under specific stochastic markets. Furthermore, in stationary and ergodic markets, the so-called Bayesian strategy utilizing true conditional distributions, based on observed past market information during investment, almost surely does not perform better than the proposed strategies in terms of empirical utility, Sharpe ratio, or growth rate, which, in contrast, do not rely on conditional distributions. | [
"['Duy Khanh Lam']"
] |
null | null | 2406.13487 | null | null | http://arxiv.org/pdf/2406.13487v1 | 2024-06-19T12:14:45Z | 2024-06-19T12:14:45Z | An evidential time-to-event prediction model based on Gaussian random
fuzzy numbers | We introduce an evidential model for time-to-event prediction with censored data. In this model, uncertainty on event time is quantified by Gaussian random fuzzy numbers, a newly introduced family of random fuzzy subsets of the real line with associated belief functions, generalizing both Gaussian random variables and Gaussian possibility distributions. Our approach makes minimal assumptions about the underlying time-to-event distribution. The model is fit by minimizing a generalized negative log-likelihood function that accounts for both normal and censored data. Comparative experiments on two real-world datasets demonstrate the very good performance of our model as compared to the state-of-the-art. | [
"['Ling Huang' 'Yucheng Xing' 'Thierry Denoeux' 'Mengling Feng']"
] |
null | null | 2406.13488 | null | null | http://arxiv.org/pdf/2406.13488v1 | 2024-06-19T12:17:14Z | 2024-06-19T12:17:14Z | Approximately Equivariant Neural Processes | Equivariant deep learning architectures exploit symmetries in learning problems to improve the sample efficiency of neural-network-based models and their ability to generalise. However, when modelling real-world data, learning problems are often not exactly equivariant, but only approximately. For example, when estimating the global temperature field from weather station observations, local topographical features like mountains break translation equivariance. In these scenarios, it is desirable to construct architectures that can flexibly depart from exact equivariance in a data-driven way. In this paper, we develop a general approach to achieving this using existing equivariant architectures. Our approach is agnostic to both the choice of symmetry group and model architecture, making it widely applicable. We consider the use of approximately equivariant architectures in neural processes (NPs), a popular family of meta-learning models. We demonstrate the effectiveness of our approach on a number of synthetic and real-world regression experiments, demonstrating that approximately equivariant NP models can outperform both their non-equivariant and strictly equivariant counterparts. | [
"['Matthew Ashman' 'Cristiana Diaconu' 'Adrian Weller' 'Wessel Bruinsma'\n 'Richard E. Turner']"
] |
null | null | 2406.13490 | null | null | http://arxiv.org/pdf/2406.13490v1 | 2024-06-19T12:20:29Z | 2024-06-19T12:20:29Z | The Surprising Benefits of Base Rate Neglect in Robust Aggregation | Robust aggregation integrates predictions from multiple experts without knowledge of the experts' information structures. Prior work assumes experts are Bayesian, providing predictions as perfect posteriors based on their signals. However, real-world experts often deviate systematically from Bayesian reasoning. Our work considers experts who tend to ignore the base rate. We find that a certain degree of base rate neglect helps with robust forecast aggregation. Specifically, we consider a forecast aggregation problem with two experts who each predict a binary world state after observing private signals. Unlike previous work, we model experts exhibiting base rate neglect, where they incorporate the base rate information to degree $lambdain[0,1]$, with $lambda=0$ indicating complete ignorance and $lambda=1$ perfect Bayesian updating. To evaluate aggregators' performance, we adopt Arieli et al. (2018)'s worst-case regret model, which measures the maximum regret across the set of considered information structures compared to an omniscient benchmark. Our results reveal the surprising V-shape of regret as a function of $lambda$. That is, predictions with an intermediate incorporating degree of base rate $lambda<1$ can counter-intuitively lead to lower regret than perfect Bayesian posteriors with $lambda=1$. We additionally propose a new aggregator with low regret robust to unknown $lambda$. Finally, we conduct an empirical study to test the base rate neglect model and evaluate the performance of various aggregators. | [
"['Yuqing Kong' 'Shu Wang' 'Ying Wang']"
] |
null | null | 2406.13493 | null | null | http://arxiv.org/pdf/2406.13493v1 | 2024-06-19T12:26:36Z | 2024-06-19T12:26:36Z | In-Context In-Context Learning with Transformer Neural Processes | Neural processes (NPs) are a powerful family of meta-learning models that seek to approximate the posterior predictive map of the ground-truth stochastic process from which each dataset in a meta-dataset is sampled. There are many cases in which practitioners, besides having access to the dataset of interest, may also have access to other datasets that share similarities with it. In this case, integrating these datasets into the NP can improve predictions. We equip NPs with this functionality and describe this paradigm as in-context in-context learning. Standard NP architectures, such as the convolutional conditional NP (ConvCNP) or the family of transformer neural processes (TNPs), are not capable of in-context in-context learning, as they are only able to condition on a single dataset. We address this shortcoming by developing the in-context in-context learning pseudo-token TNP (ICICL-TNP). The ICICL-TNP builds on the family of PT-TNPs, which utilise pseudo-token-based transformer architectures to sidestep the quadratic computational complexity associated with regular transformer architectures. Importantly, the ICICL-TNP is capable of conditioning on both sets of datapoints and sets of datasets, enabling it to perform in-context in-context learning. We demonstrate the importance of in-context in-context learning and the effectiveness of the ICICL-TNP in a number of experiments. | [
"['Matthew Ashman' 'Cristiana Diaconu' 'Adrian Weller' 'Richard E. Turner']"
] |
null | null | 2406.13499 | null | null | http://arxiv.org/pdf/2406.13499v1 | 2024-06-19T12:41:15Z | 2024-06-19T12:41:15Z | GraphMU: Repairing Robustness of Graph Neural Networks via Machine
Unlearning | Graph Neural Networks (GNNs) have demonstrated significant application potential in various fields. However, GNNs are still vulnerable to adversarial attacks. Numerous adversarial defense methods on GNNs are proposed to address the problem of adversarial attacks. However, these methods can only serve as a defense before poisoning, but cannot repair poisoned GNN. Therefore, there is an urgent need for a method to repair poisoned GNN. In this paper, we address this gap by introducing the novel concept of model repair for GNNs. We propose a repair framework, Repairing Robustness of Graph Neural Networks via Machine Unlearning (GraphMU), which aims to fine-tune poisoned GNN to forget adversarial samples without the need for complete retraining. We also introduce a unlearning validation method to ensure that our approach effectively forget specified poisoned data. To evaluate the effectiveness of GraphMU, we explore three fine-tuned subgraph construction scenarios based on the available perturbation information: (i) Known Perturbation Ratios, (ii) Known Complete Knowledge of Perturbations, and (iii) Unknown any Knowledge of Perturbations. Our extensive experiments, conducted across four citation datasets and four adversarial attack scenarios, demonstrate that GraphMU can effectively restore the performance of poisoned GNN. | [
"['Tao Wu' 'Xinwen Cao' 'Chao Wang' 'Shaojie Qiao' 'Xingping Xian'\n 'Lin Yuan' 'Canyixing Cui' 'Yanbing Liu']"
] |
null | null | 2406.13507 | null | null | http://arxiv.org/pdf/2406.13507v1 | 2024-06-19T12:54:03Z | 2024-06-19T12:54:03Z | Scalable unsupervised alignment of general metric and non-metric
structures | Aligning data from different domains is a fundamental problem in machine learning with broad applications across very different areas, most notably aligning experimental readouts in single-cell multiomics. Mathematically, this problem can be formulated as the minimization of disagreement of pair-wise quantities such as distances and is related to the Gromov-Hausdorff and Gromov-Wasserstein distances. Computationally, it is a quadratic assignment problem (QAP) that is known to be NP-hard. Prior works attempted to solve the QAP directly with entropic or low-rank regularization on the permutation, which is computationally tractable only for modestly-sized inputs, and encode only limited inductive bias related to the domains being aligned. We consider the alignment of metric structures formulated as a discrete Gromov-Wasserstein problem and instead of solving the QAP directly, we propose to learn a related well-scalable linear assignment problem (LAP) whose solution is also a minimizer of the QAP. We also show a flexible extension of the proposed framework to general non-metric dissimilarities through differentiable ranks. We extensively evaluate our approach on synthetic and real datasets from single-cell multiomics and neural latent spaces, achieving state-of-the-art performance while being conceptually and computationally simple. | [
"['Sanketh Vedula' 'Valentino Maiorca' 'Lorenzo Basile'\n 'Francesco Locatello' 'Alex Bronstein']"
] |
null | null | 2406.13533 | null | null | http://arxiv.org/pdf/2406.13533v1 | 2024-06-19T13:17:28Z | 2024-06-19T13:17:28Z | DRACO: Decentralized Asynchronous Federated Learning over Continuous
Row-Stochastic Network Matrices | Recent developments and emerging use cases, such as smart Internet of Things (IoT) and Edge AI, have sparked considerable interest in the training of neural networks over fully decentralized (serverless) networks. One of the major challenges of decentralized learning is to ensure stable convergence without resorting to strong assumptions applied for each agent regarding data distributions or updating policies. To address these issues, we propose DRACO, a novel method for decentralized asynchronous Stochastic Gradient Descent (SGD) over row-stochastic gossip wireless networks by leveraging continuous communication. Our approach enables edge devices within decentralized networks to perform local training and model exchanging along a continuous timeline, thereby eliminating the necessity for synchronized timing. The algorithm also features a specific technique of decoupling communication and computation schedules, which empowers complete autonomy for all users and manageable instructions for stragglers. Through a comprehensive convergence analysis, we highlight the advantages of asynchronous and autonomous participation in decentralized optimization. Our numerical experiments corroborate the efficacy of the proposed technique. | [
"['Eunjeong Jeong' 'Marios Kountouris']"
] |
null | null | 2406.13542 | null | null | http://arxiv.org/pdf/2406.13542v1 | 2024-06-19T13:29:53Z | 2024-06-19T13:29:53Z | Self-play with Execution Feedback: Improving Instruction-following
Capabilities of Large Language Models | One core capability of large language models (LLMs) is to follow natural language instructions. However, the issue of automatically constructing high-quality training data to enhance the complex instruction-following abilities of LLMs without manual annotation remains unresolved. In this paper, we introduce AutoIF, the first scalable and reliable method for automatically generating instruction-following training data. AutoIF transforms the validation of instruction-following data quality into code verification, requiring LLMs to generate instructions, the corresponding code to check the correctness of the instruction responses, and unit test samples to verify the code's correctness. Then, execution feedback-based rejection sampling can generate data for Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) training. AutoIF achieves significant improvements across three training algorithms, SFT, Offline DPO, and Online DPO, when applied to the top open-source LLMs, Qwen2 and LLaMA3, in self-alignment and strong-to-weak distillation settings. Our code is publicly available at https://github.com/QwenLM/AutoIF. | [
"['Guanting Dong' 'Keming Lu' 'Chengpeng Li' 'Tingyu Xia' 'Bowen Yu'\n 'Chang Zhou' 'Jingren Zhou']"
] |
null | null | 2406.13544 | null | null | http://arxiv.org/pdf/2406.13544v2 | 2024-07-03T02:53:47Z | 2024-06-19T13:30:17Z | One Fits All: Learning Fair Graph Neural Networks for Various Sensitive
Attributes | Recent studies have highlighted fairness issues in Graph Neural Networks (GNNs), where they produce discriminatory predictions against specific protected groups categorized by sensitive attributes such as race and age. While various efforts to enhance GNN fairness have made significant progress, these approaches are often tailored to specific sensitive attributes. Consequently, they necessitate retraining the model from scratch to accommodate changes in the sensitive attribute requirement, resulting in high computational costs. To gain deeper insights into this issue, we approach the graph fairness problem from a causal modeling perspective, where we identify the confounding effect induced by the sensitive attribute as the underlying reason. Motivated by this observation, we formulate the fairness problem in graphs from an invariant learning perspective, which aims to learn invariant representations across environments. Accordingly, we propose a graph fairness framework based on invariant learning, namely FairINV, which enables the training of fair GNNs to accommodate various sensitive attributes within a single training session. Specifically, FairINV incorporates sensitive attribute partition and trains fair GNNs by eliminating spurious correlations between the label and various sensitive attributes. Experimental results on several real-world datasets demonstrate that FairINV significantly outperforms state-of-the-art fairness approaches, underscoring its effectiveness. Our code is available via: https://github.com/ZzoomD/FairINV/. | [
"['Yuchang Zhu' 'Jintang Li' 'Yatao Bian' 'Zibin Zheng' 'Liang Chen']"
] |
null | null | 2406.13547 | null | null | http://arxiv.org/pdf/2406.13547v1 | 2024-06-19T13:32:47Z | 2024-06-19T13:32:47Z | ModSec-Learn: Boosting ModSecurity with Machine Learning | ModSecurity is widely recognized as the standard open-source Web Application Firewall (WAF), maintained by the OWASP Foundation. It detects malicious requests by matching them against the Core Rule Set (CRS), identifying well-known attack patterns. Each rule is manually assigned a weight based on the severity of the corresponding attack, and a request is blocked if the sum of the weights of matched rules exceeds a given threshold. However, we argue that this strategy is largely ineffective against web attacks, as detection is only based on heuristics and not customized on the application to protect. In this work, we overcome this issue by proposing a machine-learning model that uses the CRS rules as input features. Through training, ModSec-Learn is able to tune the contribution of each CRS rule to predictions, thus adapting the severity level to the web applications to protect. Our experiments show that ModSec-Learn achieves a significantly better trade-off between detection and false positive rates. Finally, we analyze how sparse regularization can reduce the number of rules that are relevant at inference time, by discarding more than 30% of the CRS rules. We release our open-source code and the dataset at https://github.com/pralab/modsec-learn and https://github.com/pralab/http-traffic-dataset, respectively. | [
"['Christian Scano' 'Giuseppe Floris' 'Biagio Montaruli' 'Luca Demetrio'\n 'Andrea Valenza' 'Luca Compagna' 'Davide Ariu' 'Luca Piras'\n 'Davide Balzarotti' 'Battista Biggio']"
] |
null | null | 2406.13552 | null | null | http://arxiv.org/pdf/2406.13552v1 | 2024-06-19T13:39:05Z | 2024-06-19T13:39:05Z | Standardness Fogs Meaning: A Position Regarding the Informed Usage of
Standard Datasets | Standard datasets are frequently used to train and evaluate Machine Learning models. However, the assumed standardness of these datasets leads to a lack of in-depth discussion on how their labels match the derived categories for the respective use case. In other words, the standardness of the datasets seems to fog coherency and applicability, thus impeding the trust in Machine Learning models. We propose to adopt Grounded Theory and Hypotheses Testing through Visualization as methods to evaluate the match between use case, derived categories, and labels of standard datasets. To showcase the approach, we apply it to the 20 Newsgroups dataset and the MNIST dataset. For the 20 Newsgroups dataset, we demonstrate that the labels are imprecise. Therefore, we argue that neither a Machine Learning model can learn a meaningful abstraction of derived categories nor one can draw conclusions from achieving high accuracy. For the MNIST dataset, we demonstrate how the labels can be confirmed to be defined well. We conclude that a concept of standardness of a dataset implies that there is a match between use case, derived categories, and class labels, as in the case of the MNIST dataset. We argue that this is necessary to learn a meaningful abstraction and, thus, improve trust in the Machine Learning model. | [
"['Tim Cech' 'Ole Wegen' 'Daniel Atzberger' 'Rico Richter' 'Willy Scheibel'\n 'Jürgen Döllner']"
] |
null | null | 2406.13559 | null | null | http://arxiv.org/pdf/2406.13559v1 | 2024-06-19T13:47:05Z | 2024-06-19T13:47:05Z | Solarcast-ML: Per Node GraphCast Extension for Solar Energy Production | This project presents an extension to the GraphCast model, a state-of-the-art graph neural network (GNN) for global weather forecasting, by integrating solar energy production forecasting capabilities. The proposed approach leverages the weather forecasts generated by GraphCast and trains a neural network model to predict the ratio of actual solar output to potential solar output based on various weather conditions. The model architecture consists of an input layer corresponding to weather features (temperature, humidity, dew point, wind speed, rain, barometric pressure, and altitude), two hidden layers with ReLU activations, and an output layer predicting solar radiation. The model is trained using a mean absolute error loss function and Adam optimizer. The results demonstrate the model's effectiveness in accurately predicting solar radiation, with its convergence behavior, decreasing training loss, and accurate prediction of solar radiation patterns suggesting successful learning of the underlying relationships between weather conditions and solar radiation. The integration of solar energy production forecasting with GraphCast offers valuable insights for the renewable energy sector, enabling better planning and decision-making based on expected solar energy production. Future work could explore further model refinements, incorporation of additional weather variables, and extension to other renewable energy sources. | [
"['Cale Colony' 'Razan Andigani']"
] |
null | null | 2406.13569 | null | null | http://arxiv.org/pdf/2406.13569v1 | 2024-06-19T13:58:42Z | 2024-06-19T13:58:42Z | Bayes' capacity as a measure for reconstruction attacks in federated
learning | Within the machine learning community, reconstruction attacks are a principal attack of concern and have been identified even in federated learning, which was designed with privacy preservation in mind. In federated learning, it has been shown that an adversary with knowledge of the machine learning architecture is able to infer the exact value of a training element given an observation of the weight updates performed during stochastic gradient descent. In response to these threats, the privacy community recommends the use of differential privacy in the stochastic gradient descent algorithm, termed DP-SGD. However, DP has not yet been formally established as an effective countermeasure against reconstruction attacks. In this paper, we formalise the reconstruction threat model using the information-theoretic framework of quantitative information flow. We show that the Bayes' capacity, related to the Sibson mutual information of order infinity, represents a tight upper bound on the leakage of the DP-SGD algorithm to an adversary interested in performing a reconstruction attack. We provide empirical results demonstrating the effectiveness of this measure for comparing mechanisms against reconstruction threats. | [
"['Sayan Biswas' 'Mark Dras' 'Pedro Faustini' 'Natasha Fernandes'\n 'Annabelle McIver' 'Catuscia Palamidessi' 'Parastoo Sadeghi']"
] |
null | null | 2406.13584 | null | null | http://arxiv.org/pdf/2406.13584v1 | 2024-06-19T14:19:59Z | 2024-06-19T14:19:59Z | Explaining time series models using frequency masking | Time series data is fundamentally important for describing many critical domains such as healthcare, finance, and climate, where explainable models are necessary for safe automated decision-making. To develop eXplainable AI (XAI) in these domains therefore implies explaining salient information in the time series. Current methods for obtaining saliency maps assumes localized information in the raw input space. In this paper, we argue that the salient information of a number of time series is more likely to be localized in the frequency domain. We propose FreqRISE, which uses masking based methods to produce explanations in the frequency and time-frequency domain, which shows the best performance across a number of tasks. | [
"['Thea Brüsch' 'Kristoffer K. Wickstrøm' 'Mikkel N. Schmidt'\n 'Tommy S. Alstrøm' 'Robert Jenssen']"
] |
null | null | 2406.13597 | null | null | http://arxiv.org/pdf/2406.13597v1 | 2024-06-19T14:41:09Z | 2024-06-19T14:41:09Z | GraphKAN: Enhancing Feature Extraction with Graph Kolmogorov Arnold
Networks | Massive number of applications involve data with underlying relationships embedded in non-Euclidean space. Graph neural networks (GNNs) are utilized to extract features by capturing the dependencies within graphs. Despite groundbreaking performances, we argue that Multi-layer perceptrons (MLPs) and fixed activation functions impede the feature extraction due to information loss. Inspired by Kolmogorov Arnold Networks (KANs), we make the first attempt to GNNs with KANs. We discard MLPs and activation functions, and instead used KANs for feature extraction. Experiments demonstrate the effectiveness of GraphKAN, emphasizing the potential of KANs as a powerful tool. Code is available at https://github.com/Ryanfzhang/GraphKan. | [
"['Fan Zhang' 'Xin Zhang']"
] |
null | null | 2406.13619 | null | null | http://arxiv.org/pdf/2406.13619v2 | 2024-07-14T05:54:39Z | 2024-06-19T15:15:00Z | Generative Modeling by Minimizing the Wasserstein-2 Loss | This paper approaches the unsupervised learning problem by minimizing the second-order Wasserstein loss (the $W_2$ loss) through a distribution-dependent ordinary differential equation (ODE), whose dynamics involves the Kantorovich potential associated with the true data distribution and a current estimate of it. A main result shows that the time-marginal laws of the ODE form a gradient flow for the $W_2$ loss, which converges exponentially to the true data distribution. An Euler scheme for the ODE is proposed and it is shown to recover the gradient flow for the $W_2$ loss in the limit. An algorithm is designed by following the scheme and applying persistent training, which naturally fits our gradient-flow approach. In both low- and high-dimensional experiments, our algorithm outperforms Wasserstein generative adversarial networks by increasing the level of persistent training appropriately. | [
"['Yu-Jui Huang' 'Zachariah Malik']"
] |
null | null | 2406.13621 | null | null | http://arxiv.org/pdf/2406.13621v1 | 2024-06-19T15:17:10Z | 2024-06-19T15:17:10Z | Improving Visual Commonsense in Language Models via Multiple Image
Generation | Commonsense reasoning is fundamentally based on multimodal knowledge. However, existing large language models (LLMs) are primarily trained using textual data only, limiting their ability to incorporate essential visual information. In contrast, Visual Language Models, which excel at visually-oriented tasks, often fail at non-visual tasks such as basic commonsense reasoning. This divergence highlights a critical challenge - the integration of robust visual understanding with foundational text-based language reasoning. To this end, we introduce a method aimed at enhancing LLMs' visual commonsense. Specifically, our method generates multiple images based on the input text prompt and integrates these into the model's decision-making process by mixing their prediction probabilities. To facilitate multimodal grounded language modeling, we employ a late-fusion layer that combines the projected visual features with the output of a pre-trained LLM conditioned on text only. This late-fusion layer enables predictions based on comprehensive image-text knowledge as well as text only when this is required. We evaluate our approach using several visual commonsense reasoning tasks together with traditional NLP tasks, including common sense reasoning and reading comprehension. Our experimental results demonstrate significant superiority over existing baselines. When applied to recent state-of-the-art LLMs (e.g., Llama3), we observe improvements not only in visual common sense but also in traditional NLP benchmarks. Code and models are available under https://github.com/guyyariv/vLMIG. | [
"['Guy Yariv' 'Idan Schwartz' 'Yossi Adi' 'Sagie Benaim']"
] |
null | null | 2406.13627 | null | null | http://arxiv.org/pdf/2406.13627v1 | 2024-06-19T15:20:28Z | 2024-06-19T15:20:28Z | Can AI be enabled to dynamical downscaling? Training a Latent Diffusion
Model to mimic km-scale COSMO-CLM downscaling of ERA5 over Italy | Downscaling techniques are one of the most prominent applications of Deep Learning (DL) in Earth System Modeling. A robust DL downscaling model can generate high-resolution fields from coarse-scale numerical model simulations, saving the timely and resourceful applications of regional/local models. Additionally, generative DL models have the potential to provide uncertainty information, by generating ensemble-like scenario pools, a task that is computationally prohibitive for traditional numerical simulations. In this study, we apply a Latent Diffusion Model (LDM) to downscale ERA5 data over Italy up to a resolution of 2 km. The high-resolution target data consists of results from a high-resolution dynamical downscaling performed with COSMO-CLM. Our goal is to demonstrate that recent advancements in generative modeling enable DL-based models to deliver results comparable to those of numerical dynamical downscaling models, given the same input data (i.e., ERA5 data), preserving the realism of fine-scale features and flow characteristics. The training and testing database consists of hourly data from 2000 to 2020. The target variables of this study are 2-m temperature and 10-m horizontal wind components. A selection of predictors from ERA5 is used as input to the LDM, and a residual approach against a reference UNET is leveraged in applying the LDM. The performance of the generative LDM is compared with reference baselines of increasing complexity: quadratic interpolation of ERA5, a UNET, and a Generative Adversarial Network (GAN) built on the same reference UNET. Results highlight the improvements introduced by the LDM architecture and the residual approach over these baselines. The models are evaluated on a yearly test dataset, assessing the models' performance through deterministic metrics, spatial distribution of errors, and reconstruction of frequency and power spectra distributions. | [
"['Elena Tomasi' 'Gabriele Franch' 'Marco Cristoforetti']"
] |
null | null | 2406.13629 | null | null | http://arxiv.org/pdf/2406.13629v1 | 2024-06-19T15:25:29Z | 2024-06-19T15:25:29Z | InstructRAG: Instructing Retrieval-Augmented Generation with Explicit
Denoising | Retrieval-augmented generation (RAG) has shown promising potential to enhance the accuracy and factuality of language models (LMs). However, imperfect retrievers or noisy corpora can introduce misleading or even erroneous information to the retrieved contents, posing a significant challenge to the generation quality. Existing RAG methods typically address this challenge by directly predicting final answers despite potentially noisy inputs, resulting in an implicit denoising process that is difficult to interpret and verify. On the other hand, the acquisition of explicit denoising supervision is often costly, involving significant human efforts. In this work, we propose InstructRAG, where LMs explicitly learn the denoising process through self-synthesized rationales -- First, we instruct the LM to explain how the ground-truth answer is derived from retrieved documents. Then, these rationales can be used either as demonstrations for in-context learning of explicit denoising or as supervised fine-tuning data to train the model. Compared to standard RAG approaches, InstructRAG requires no additional supervision, allows for easier verification of the predicted answers, and effectively improves generation accuracy. Experiments show InstructRAG consistently outperforms existing RAG methods in both training-free and trainable scenarios, achieving a relative improvement of 8.3% over the best baseline method on average across five knowledge-intensive benchmarks. Extensive analysis indicates that InstructRAG scales well with increased numbers of retrieved documents and consistently exhibits robust denoising ability even in out-of-domain datasets, demonstrating strong generalizability. | [
"['Zhepei Wei' 'Wei-Lin Chen' 'Yu Meng']"
] |
null | null | 2406.13633 | null | null | http://arxiv.org/pdf/2406.13633v1 | 2024-06-19T15:29:14Z | 2024-06-19T15:29:14Z | Reinforcement Learning for Infinite-Horizon Average-Reward MDPs with
Multinomial Logistic Function Approximation | We study model-based reinforcement learning with non-linear function approximation where the transition function of the underlying Markov decision process (MDP) is given by a multinomial logistic (MNL) model. In this paper, we develop two algorithms for the infinite-horizon average reward setting. Our first algorithm texttt{UCRL2-MNL} applies to the class of communicating MDPs and achieves an $tilde{mathcal{O}}(dDsqrt{T})$ regret, where $d$ is the dimension of feature mapping, $D$ is the diameter of the underlying MDP, and $T$ is the horizon. The second algorithm texttt{OVIFH-MNL} is computationally more efficient and applies to the more general class of weakly communicating MDPs, for which we show a regret guarantee of $tilde{mathcal{O}}(d^{2/5} mathrm{sp}(v^*)T^{4/5})$ where $mathrm{sp}(v^*)$ is the span of the associated optimal bias function. We also prove a lower bound of $Omega(dsqrt{DT})$ for learning communicating MDPs with MNL transitions of diameter at most $D$. Furthermore, we show a regret lower bound of $Omega(dH^{3/2}sqrt{K})$ for learning $H$-horizon episodic MDPs with MNL function approximation where $K$ is the number of episodes, which improves upon the best-known lower bound for the finite-horizon setting. | [
"['Jaehyun Park' 'Dabeen Lee']"
] |
null | null | 2406.13636 | null | null | http://arxiv.org/pdf/2406.13636v1 | 2024-06-19T15:31:21Z | 2024-06-19T15:31:21Z | Contrast Sets for Evaluating Language-Guided Robot Policies | Robot evaluations in language-guided, real world settings are time-consuming and often sample only a small space of potential instructions across complex scenes. In this work, we introduce contrast sets for robotics as an approach to make small, but specific, perturbations to otherwise independent, identically distributed (i.i.d.) test instances. We investigate the relationship between experimenter effort to carry out an evaluation and the resulting estimated test performance as well as the insights that can be drawn from performance on perturbed instances. We use contrast sets to characterize policies at reduced experimenter effort in both a simulated manipulation task and a physical robot vision-and-language navigation task. We encourage the use of contrast set evaluations as a more informative alternative to small scale, i.i.d. demonstrations on physical robots, and as a scalable alternative to industry-scale real world evaluations. | [
"['Abrar Anwar' 'Rohan Gupta' 'Jesse Thomason']"
] |
null | null | 2406.13640 | null | null | http://arxiv.org/pdf/2406.13640v2 | 2024-07-15T04:17:11Z | 2024-06-19T15:39:27Z | Transferable Tactile Transformers for Representation Learning Across
Diverse Sensors and Tasks | This paper presents T3: Transferable Tactile Transformers, a framework for tactile representation learning that scales across multi-sensors and multi-tasks. T3 is designed to overcome the contemporary issue that camera-based tactile sensing is extremely heterogeneous, i.e. sensors are built into different form factors, and existing datasets were collected for disparate tasks. T3 captures the shared latent information across different sensor-task pairings by constructing a shared trunk transformer with sensor-specific encoders and task-specific decoders. The pre-training of T3 utilizes a novel Foundation Tactile (FoTa) dataset, which is aggregated from several open-sourced datasets and it contains over 3 million data points gathered from 13 sensors and 11 tasks. FoTa is the largest and most diverse dataset in tactile sensing to date and it is made publicly available in a unified format. Across various sensors and tasks, experiments show that T3 pre-trained with FoTa achieved zero-shot transferability in certain sensor-task pairings, can be further fine-tuned with small amounts of domain-specific data, and its performance scales with bigger network sizes. T3 is also effective as a tactile encoder for long horizon contact-rich manipulation. Results from sub-millimeter multi-pin electronics insertion tasks show that T3 achieved a task success rate 25% higher than that of policies trained with tactile encoders trained from scratch, or 53% higher than without tactile sensing. Data, code, and model checkpoints are open-sourced at https://t3.alanz.info. | [
"['Jialiang Zhao' 'Yuxiang Ma' 'Lirui Wang' 'Edward H. Adelson']"
] |
null | null | 2406.13653 | null | null | http://arxiv.org/pdf/2406.13653v1 | 2024-06-19T15:56:21Z | 2024-06-19T15:56:21Z | Controlling Forgetting with Test-Time Data in Continual Learning | Foundational vision-language models have shown impressive performance on various downstream tasks. Yet, there is still a pressing need to update these models later as new tasks or domains become available. Ongoing Continual Learning (CL) research provides techniques to overcome catastrophic forgetting of previous information when new knowledge is acquired. To date, CL techniques focus only on the supervised training sessions. This results in significant forgetting yielding inferior performance to even the prior model zero shot performance. In this work, we argue that test-time data hold great information that can be leveraged in a self supervised manner to refresh the model's memory of previous learned tasks and hence greatly reduce forgetting at no extra labelling cost. We study how unsupervised data can be employed online to improve models' performance on prior tasks upon encountering representative samples. We propose a simple yet effective student-teacher model with gradient based sparse parameters updates and show significant performance improvements and reduction in forgetting, which could alleviate the role of an offline episodic memory/experience replay buffer. | [
"['Vaibhav Singh' 'Rahaf Aljundi' 'Eugene Belilovsky']"
] |
null | null | 2406.13655 | null | null | http://arxiv.org/pdf/2406.13655v1 | 2024-06-19T15:58:35Z | 2024-06-19T15:58:35Z | Improving GFlowNets with Monte Carlo Tree Search | Generative Flow Networks (GFlowNets) treat sampling from distributions over compositional discrete spaces as a sequential decision-making problem, training a stochastic policy to construct objects step by step. Recent studies have revealed strong connections between GFlowNets and entropy-regularized reinforcement learning. Building on these insights, we propose to enhance planning capabilities of GFlowNets by applying Monte Carlo Tree Search (MCTS). Specifically, we show how the MENTS algorithm (Xiao et al., 2019) can be adapted for GFlowNets and used during both training and inference. Our experiments demonstrate that this approach improves the sample efficiency of GFlowNet training and the generation fidelity of pre-trained GFlowNet models. | [
"['Nikita Morozov' 'Daniil Tiapkin' 'Sergey Samsonov' 'Alexey Naumov'\n 'Dmitry Vetrov']"
] |
null | null | 2406.13661 | null | null | http://arxiv.org/pdf/2406.13661v1 | 2024-06-19T16:08:00Z | 2024-06-19T16:08:00Z | Hitchhiker's guide on Energy-Based Models: a comprehensive review on the
relation with other generative models, sampling and statistical physics | Energy-Based Models (EBMs) have emerged as a powerful framework in the realm of generative modeling, offering a unique perspective that aligns closely with principles of statistical mechanics. This review aims to provide physicists with a comprehensive understanding of EBMs, delineating their connection to other generative models such as Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and Normalizing Flows. We explore the sampling techniques crucial for EBMs, including Markov Chain Monte Carlo (MCMC) methods, and draw parallels between EBM concepts and statistical mechanics, highlighting the significance of energy functions and partition functions. Furthermore, we delve into state-of-the-art training methodologies for EBMs, covering recent advancements and their implications for enhanced model performance and efficiency. This review is designed to clarify the often complex interconnections between these models, which can be challenging due to the diverse communities working on the topic. | [
"['Davide Carbone']"
] |
null | null | 2406.13663 | null | null | http://arxiv.org/pdf/2406.13663v2 | 2024-07-01T12:39:26Z | 2024-06-19T16:10:26Z | Model Internals-based Answer Attribution for Trustworthy
Retrieval-Augmented Generation | Ensuring the verifiability of model answers is a fundamental challenge for retrieval-augmented generation (RAG) in the question answering (QA) domain. Recently, self-citation prompting was proposed to make large language models (LLMs) generate citations to supporting documents along with their answers. However, self-citing LLMs often struggle to match the required format, refer to non-existent sources, and fail to faithfully reflect LLMs' context usage throughout the generation. In this work, we present MIRAGE --Model Internals-based RAG Explanations -- a plug-and-play approach using model internals for faithful answer attribution in RAG applications. MIRAGE detects context-sensitive answer tokens and pairs them with retrieved documents contributing to their prediction via saliency methods. We evaluate our proposed approach on a multilingual extractive QA dataset, finding high agreement with human answer attribution. On open-ended QA, MIRAGE achieves citation quality and efficiency comparable to self-citation while also allowing for a finer-grained control of attribution parameters. Our qualitative evaluation highlights the faithfulness of MIRAGE's attributions and underscores the promising application of model internals for RAG answer attribution. | [
"['Jirui Qi' 'Gabriele Sarti' 'Raquel Fernández' 'Arianna Bisazza']"
] |
null | null | 2406.13665 | null | null | http://arxiv.org/pdf/2406.13665v1 | 2024-06-19T16:11:59Z | 2024-06-19T16:11:59Z | Challenges in Binary Classification | Binary Classification plays an important role in machine learning. For linear classification, SVM is the optimal binary classification method. For nonlinear classification, the SVM algorithm needs to complete the classification task by using the kernel function. Although the SVM algorithm with kernel function is very effective, the selection of kernel function is empirical, which means that the kernel function may not be optimal. Therefore, it is worth studying how to obtain an optimal binary classifier. In this paper, the problem of finding the optimal binary classifier is considered as a variational problem. We design the objective function of this variational problem through the max-min problem of the (Euclidean) distance between two classes. For linear classification, it can be deduced that SVM is a special case of this variational problem framework. For Euclidean distance, it is proved that the proposed variational problem has some limitations for nonlinear classification. Therefore, how to design a more appropriate objective function to find the optimal binary classifier is still an open problem. Further, it's discussed some challenges and problems in finding the optimal classifier. | [
"['Pengbo Yang' 'Jian Yu']"
] |
null | null | 2406.13668 | null | null | http://arxiv.org/pdf/2406.13668v1 | 2024-06-19T16:19:39Z | 2024-06-19T16:19:39Z | Improved bounds for calibration via stronger sign preservation games | A set of probabilistic forecasts is calibrated if each prediction of the forecaster closely approximates the empirical distribution of outcomes on the subset of timesteps where that prediction was made. We study the fundamental problem of online calibrated forecasting of binary sequences, which was initially studied by Foster & Vohra (1998). They derived an algorithm with $O(T^{2/3})$ calibration error after $T$ time steps, and showed a lower bound of $Omega(T^{1/2})$. These bounds remained stagnant for two decades, until Qiao & Valiant (2021) improved the lower bound to $Omega(T^{0.528})$ by introducing a combinatorial game called sign preservation and showing that lower bounds for this game imply lower bounds for calibration. We introduce a strengthening of Qiao & Valiant's game that we call sign preservation with reuse (SPR). We prove that the relationship between SPR and calibrated forecasting is bidirectional: not only do lower bounds for SPR translate into lower bounds for calibration, but algorithms for SPR also translate into new algorithms for calibrated forecasting. In particular, any strategy that improves the trivial upper bound for the value of the SPR game would imply a forecasting algorithm with calibration error exponent less than 2/3, improving Foster & Vohra's upper bound for the first time. Using similar ideas, we then prove a slightly stronger lower bound than that of Qiao & Valiant, namely $Omega(T^{0.54389})$. Our lower bound is obtained by an oblivious adversary, marking the first $omega(T^{1/2})$ calibration lower bound for oblivious adversaries. | [
"['Yuval Dagan' 'Constantinos Daskalakis' 'Maxwell Fishelson'\n 'Noah Golowich' 'Robert Kleinberg' 'Princewill Okoroafor']"
] |
null | null | 2406.13679 | null | null | http://arxiv.org/pdf/2406.13679v1 | 2024-06-19T16:32:27Z | 2024-06-19T16:32:27Z | Prose-to-P4: Leveraging High Level Languages | Languages such as P4 and NPL have enabled a wide and diverse range of networking applications that take advantage of programmable dataplanes. However, software development in these languages is difficult. To address this issue, high-level languages have been designed to offer programmers powerful abstractions that reduce the time, effort and domain-knowledge required for developing networking applications. These languages are then translated by a compiler into P4/NPL code. Inspired by the recent success of Large Language Models (LLMs) in the task of code generation, we propose to raise the level of abstraction even higher, employing LLMs to translate prose into high-level networking code. We analyze the problem, focusing on the motivation and opportunities, as well as the challenges involved and sketch out a roadmap for the development of a system that can generate high-level dataplane code from natural language instructions. We present some promising preliminary results on generating Lucid code from natural language. | [
"['Mihai-Valentin Dumitru' 'Vlad-Andrei Bădoiu' 'Costin Raiciu']"
] |
null | null | 2406.13681 | null | null | http://arxiv.org/pdf/2406.13681v1 | 2024-06-19T16:35:23Z | 2024-06-19T16:35:23Z | On the Consistency of Fairness Measurement Methods for Regression Tasks | With growing applications of Machine Learning (ML) techniques in the real world, it is highly important to ensure that these models work in an equitable manner. One main step in ensuring fairness is to effectively measure fairness, and to this end, various metrics have been proposed in the past literature. While the computation of those metrics are straightforward in the classification set-up, it is computationally intractable in the regression domain. To address the challenge of computational intractability, past literature proposed various methods to approximate such metrics. However, they did not verify the extent to which the output of such approximation algorithms are consistent with each other. To fill this gap, this paper comprehensively studies the consistency of the output of various fairness measurement methods through conducting an extensive set of experiments on various regression tasks. As a result, it finds that while some fairness measurement approaches show strong consistency across various regression tasks, certain methods show a relatively poor consistency in certain regression tasks. This, in turn, calls for a more principled approach for measuring fairness in the regression domain. | [
"['Abdalwahab Almajed' 'Maryam Tabar' 'Peyman Najafirad']"
] |
null | null | 2406.13720 | null | null | http://arxiv.org/pdf/2406.13720v1 | 2024-06-19T17:24:36Z | 2024-06-19T17:24:36Z | On the Utility of Domain-Adjacent Fine-Tuned Model Ensembles for
Few-shot Problems | Large Language Models (LLMs) have been observed to perform well on a wide range of downstream tasks when fine-tuned on domain-specific data. However, such data may not be readily available in many applications, motivating zero-shot or few-shot approaches using domain-adjacent models. While several fine-tuned models for various tasks are available, finding an appropriate domain-adjacent model for a given task is often not straight forward. In this paper, we study DAFT-E, a framework that utilizes an Ensemble of Domain-Adjacent Fine-Tuned Foundation Models for few-shot problems. We show that for zero-shot problems, this ensembling method provides an accuracy performance close to that of the single best model. With few-shot problems, this performance improves further, at which point DEFT-E can outperform any single domain-adjacent model while requiring much less data for domain-specific fine-tuning. | [
"['Md Ibrahim Ibne Alam' 'Parikshit Ram' 'Soham Dan' 'Horst Samulowitz'\n 'Koushik Kar']"
] |
null | null | 2406.13725 | null | null | http://arxiv.org/pdf/2406.13725v1 | 2024-06-19T17:40:11Z | 2024-06-19T17:40:11Z | Tree-Sliced Wasserstein Distance on a System of Lines | Sliced Wasserstein (SW) distance in Optimal Transport (OT) is widely used in various applications thanks to its statistical effectiveness and computational efficiency. On the other hand, Tree Wassenstein (TW) and Tree-sliced Wassenstein (TSW) are instances of OT for probability measures where its ground cost is a tree metric. TSW also has a low computational complexity, i.e. linear to the number of edges in the tree. Especially, TSW is identical to SW when the tree is a chain. While SW is prone to loss of topological information of input measures due to relying on one-dimensional projection, TSW is more flexible and has a higher degree of freedom by choosing a tree rather than a line to alleviate the curse of dimensionality in SW. However, for practical applications, popular tree metric sampling methods are heavily built upon given supports, which limits their capacity to adapt to new supports. In this paper, we propose the Tree-Sliced Wasserstein distance on a System of Lines (TSW-SL), which brings a connection between SW and TSW. Compared to SW and TSW, our TSW-SL benefits from the higher degree of freedom of TSW while being suitable to dynamic settings as SW. In TSW-SL, we use a variant of the Radon Transform to project measures onto a system of lines, resulting in measures on a space with a tree metric, then leverage TW to efficiently compute distances between them. We empirically verify the advantages of TSW-SL over the traditional SW by conducting a variety of experiments on gradient flows, image style transfer, and generative models. | [
"['Viet-Hoang Tran' 'Trang Pham' 'Tho Tran' 'Tam Le' 'Tan M. Nguyen']"
] |
null | null | 2406.13726 | null | null | http://arxiv.org/pdf/2406.13726v1 | 2024-06-19T17:42:53Z | 2024-06-19T17:42:53Z | Global Solutions to Master Equations for Continuous Time Heterogeneous
Agent Macroeconomic Models | We propose and compare new global solution algorithms for continuous time heterogeneous agent economies with aggregate shocks. First, we approximate the agent distribution so that equilibrium in the economy can be characterized by a high, but finite, dimensional non-linear partial differential equation. We consider different approximations: discretizing the number of agents, discretizing the agent state variables, and projecting the distribution onto a finite set of basis functions. Second, we represent the value function using a neural network and train it to solve the differential equation using deep learning tools. We refer to the solution as an Economic Model Informed Neural Network (EMINN). The main advantage of this technique is that it allows us to find global solutions to high dimensional, non-linear problems. We demonstrate our algorithm by solving important models in the macroeconomics and spatial literatures (e.g. Krusell and Smith (1998), Khan and Thomas (2007), Bilal (2023)). | [
"['Zhouzhou Gu' 'Mathieu Laurière' 'Sebastian Merkel' 'Jonathan Payne']"
] |
null | null | 2406.13731 | null | null | http://arxiv.org/pdf/2406.13731v1 | 2024-06-19T17:54:31Z | 2024-06-19T17:54:31Z | Integrating Fuzzy Logic with Causal Inference: Enhancing the Pearl and
Neyman-Rubin Methodologies | In this paper, we generalize the Pearl and Neyman-Rubin methodologies in causal inference by introducing a generalized approach that incorporates fuzzy logic. Indeed, we introduce a fuzzy causal inference approach that consider both the vagueness and imprecision inherent in data, as well as the subjective human perspective characterized by fuzzy terms such as 'high', 'medium', and 'low'. To do so, we introduce two fuzzy causal effect formulas: the Fuzzy Average Treatment Effect (FATE) and the Generalized Fuzzy Average Treatment Effect (GFATE), together with their normalized versions: NFATE and NGFATE. When dealing with a binary treatment variable, our fuzzy causal effect formulas coincide with classical Average Treatment Effect (ATE) formula, that is a well-established and popular metric in causal inference. In FATE, all values of the treatment variable are considered equally important. In contrast, GFATE takes into account the rarity and frequency of these values. We show that for linear Structural Equation Models (SEMs), the normalized versions of our formulas, NFATE and NGFATE, are equivalent to ATE. Further, we provide identifiability criteria for these formulas and show their stability with respect to minor variations in the fuzzy subsets and the probability distributions involved. This ensures the robustness of our approach in handling small perturbations in the data. Finally, we provide several experimental examples to empirically validate and demonstrate the practical application of our proposed fuzzy causal inference methods. | [
"['Amir Saki' 'Usef Faghihi']"
] |
null | null | 2406.13733 | null | null | http://arxiv.org/pdf/2406.13733v1 | 2024-06-19T17:58:40Z | 2024-06-19T17:58:40Z | You can't handle the (dirty) truth: Data-centric insights improve
pseudo-labeling | Pseudo-labeling is a popular semi-supervised learning technique to leverage unlabeled data when labeled samples are scarce. The generation and selection of pseudo-labels heavily rely on labeled data. Existing approaches implicitly assume that the labeled data is gold standard and 'perfect'. However, this can be violated in reality with issues such as mislabeling or ambiguity. We address this overlooked aspect and show the importance of investigating labeled data quality to improve any pseudo-labeling method. Specifically, we introduce a novel data characterization and selection framework called DIPS to extend pseudo-labeling. We select useful labeled and pseudo-labeled samples via analysis of learning dynamics. We demonstrate the applicability and impact of DIPS for various pseudo-labeling methods across an extensive range of real-world tabular and image datasets. Additionally, DIPS improves data efficiency and reduces the performance distinctions between different pseudo-labelers. Overall, we highlight the significant benefits of a data-centric rethinking of pseudo-labeling in real-world settings. | [
"['Nabeel Seedat' 'Nicolas Huynh' 'Fergus Imrie' 'Mihaela van der Schaar']"
] |
null | null | 2406.13735 | null | null | http://arxiv.org/pdf/2406.13735v1 | 2024-06-19T17:59:40Z | 2024-06-19T17:59:40Z | StableSemantics: A Synthetic Language-Vision Dataset of Semantic
Representations in Naturalistic Images | Understanding the semantics of visual scenes is a fundamental challenge in Computer Vision. A key aspect of this challenge is that objects sharing similar semantic meanings or functions can exhibit striking visual differences, making accurate identification and categorization difficult. Recent advancements in text-to-image frameworks have led to models that implicitly capture natural scene statistics. These frameworks account for the visual variability of objects, as well as complex object co-occurrences and sources of noise such as diverse lighting conditions. By leveraging large-scale datasets and cross-attention conditioning, these models generate detailed and contextually rich scene representations. This capability opens new avenues for improving object recognition and scene understanding in varied and challenging environments. Our work presents StableSemantics, a dataset comprising 224 thousand human-curated prompts, processed natural language captions, over 2 million synthetic images, and 10 million attention maps corresponding to individual noun chunks. We explicitly leverage human-generated prompts that correspond to visually interesting stable diffusion generations, provide 10 generations per phrase, and extract cross-attention maps for each image. We explore the semantic distribution of generated images, examine the distribution of objects within images, and benchmark captioning and open vocabulary segmentation methods on our data. To the best of our knowledge, we are the first to release a diffusion dataset with semantic attributions. We expect our proposed dataset to catalyze advances in visual semantic understanding and provide a foundation for developing more sophisticated and effective visual models. Website: https://stablesemantics.github.io/StableSemantics | [
"['Rushikesh Zawar' 'Shaurya Dewan' 'Andrew F. Luo' 'Margaret M. Henderson'\n 'Michael J. Tarr' 'Leila Wehbe']"
] |
null | null | 2406.13743 | null | null | http://arxiv.org/pdf/2406.13743v2 | 2024-06-21T19:09:36Z | 2024-06-19T18:00:07Z | GenAI-Bench: Evaluating and Improving Compositional Text-to-Visual
Generation | While text-to-visual models now produce photo-realistic images and videos, they struggle with compositional text prompts involving attributes, relationships, and higher-order reasoning such as logic and comparison. In this work, we conduct an extensive human study on GenAI-Bench to evaluate the performance of leading image and video generation models in various aspects of compositional text-to-visual generation. We also compare automated evaluation metrics against our collected human ratings and find that VQAScore -- a metric measuring the likelihood that a VQA model views an image as accurately depicting the prompt -- significantly outperforms previous metrics such as CLIPScore. In addition, VQAScore can improve generation in a black-box manner (without finetuning) via simply ranking a few (3 to 9) candidate images. Ranking by VQAScore is 2x to 3x more effective than other scoring methods like PickScore, HPSv2, and ImageReward at improving human alignment ratings for DALL-E 3 and Stable Diffusion, especially on compositional prompts that require advanced visio-linguistic reasoning. We will release a new GenAI-Rank benchmark with over 40,000 human ratings to evaluate scoring metrics on ranking images generated from the same prompt. Lastly, we discuss promising areas for improvement in VQAScore, such as addressing fine-grained visual details. We will release all human ratings (over 80,000) to facilitate scientific benchmarking of both generative models and automated metrics. | [
"['Baiqi Li' 'Zhiqiu Lin' 'Deepak Pathak' 'Jiayao Li' 'Yixin Fei'\n 'Kewen Wu' 'Tiffany Ling' 'Xide Xia' 'Pengchuan Zhang' 'Graham Neubig'\n 'Deva Ramanan']"
] |
null | null | 2406.13748 | null | null | http://arxiv.org/pdf/2406.13748v1 | 2024-06-19T18:01:08Z | 2024-06-19T18:01:08Z | Every Language Counts: Learn and Unlearn in Multilingual LLMs | This paper investigates the propagation of harmful information in multilingual large language models (LLMs) and evaluates the efficacy of various unlearning methods. We demonstrate that fake information, regardless of the language it is in, once introduced into these models through training data, can spread across different languages, compromising the integrity and reliability of the generated content. Our findings reveal that standard unlearning techniques, which typically focus on English data, are insufficient in mitigating the spread of harmful content in multilingual contexts and could inadvertently reinforce harmful content across languages. We show that only by addressing harmful responses in both English and the original language of the harmful data can we effectively eliminate generations for all languages. This underscores the critical need for comprehensive unlearning strategies that consider the multilingual nature of modern LLMs to enhance their safety and reliability across diverse linguistic landscapes. | [
"['Taiming Lu' 'Philipp Koehn']"
] |
null | null | 2406.13750 | null | null | http://arxiv.org/pdf/2406.13750v1 | 2024-06-19T18:10:06Z | 2024-06-19T18:10:06Z | Empowering Tuberculosis Screening with Explainable Self-Supervised Deep
Neural Networks | Tuberculosis persists as a global health crisis, especially in resource-limited populations and remote regions, with more than 10 million individuals newly infected annually. It stands as a stark symbol of inequity in public health. Tuberculosis impacts roughly a quarter of the global populace, with the majority of cases concentrated in eight countries, accounting for two-thirds of all tuberculosis infections. Although a severe ailment, tuberculosis is both curable and manageable. However, early detection and screening of at-risk populations are imperative. Chest x-ray stands as the predominant imaging technique utilized in tuberculosis screening efforts. However, x-ray screening necessitates skilled radiologists, a resource often scarce, particularly in remote regions with limited resources. Consequently, there is a pressing need for artificial intelligence (AI)-powered systems to support clinicians and healthcare providers in swift screening. However, training a reliable AI model necessitates large-scale high-quality data, which can be difficult and costly to acquire. Inspired by these challenges, in this work, we introduce an explainable self-supervised self-train learning network tailored for tuberculosis case screening. The network achieves an outstanding overall accuracy of 98.14% and demonstrates high recall and precision rates of 95.72% and 99.44%, respectively, in identifying tuberculosis cases, effectively capturing clinically significant features. | [
"['Neel Patel' 'Alexander Wong' 'Ashkan Ebadi']"
] |
null | null | 2406.13754 | null | null | http://arxiv.org/pdf/2406.13754v1 | 2024-06-19T18:12:02Z | 2024-06-19T18:12:02Z | Concept Drift Visualization of SVM with Shifting Window | In machine learning, concept drift is an evolution of information that invalidates the current data model. It happens when the statistical properties of the input data change over time in unforeseen ways. Concept drift detection is crucial when dealing with dynamically changing data. Its visualization can bring valuable insight into the data dynamics, especially for multidimensional data, and is related to visual knowledge discovery. We propose a novel visualization model based on parallel coordinates, denoted as parallel histograms through time. Our model represents histograms of feature distributions for successive time-shifted windows. The drift is shown as variations of these histograms, obtained by connecting the means of the distribution for successive time windows. We show how these diagrams can be used to explain the decision made by the machine learning model in choosing the drift point. By isolating the drift at the edges of successive time windows, there will be none (or reduced) drift within the adjacent windows. We illustrate this concept on both synthetic and real datasets. In our experiments, we use an incremental/decremental SVM with shifting window, introduced by us in previous work. With our proposed technique, in addition to detect the presence of concept drift, we can also depict it. This information can be further used to explain the change. mental results, opening the possibility for further investigations. | [
"['Honorius Galmeanu' 'Razvan Andonie']"
] |
null | null | 2406.13761 | null | null | http://arxiv.org/pdf/2406.13761v1 | 2024-06-19T18:22:23Z | 2024-06-19T18:22:23Z | Exponential time differencing for matrix-valued dynamical systems | Matrix evolution equations occur in many applications, such as dynamical Lyapunov/Sylvester systems or Riccati equations in optimization and stochastic control, machine learning or data assimilation. In many cases, their tightest stability condition is coming from a linear term. Exponential time differencing (ETD) is known to produce highly stable numerical schemes by treating the linear term in an exact fashion. In particular, for stiff problems, ETD methods are a method of choice. We propose an extension of the class of ETD algorithms to matrix-valued dynamical equations. This allows us to produce highly efficient and stable integration schemes. We show their efficiency and applicability for a variety of real-world problems, from geophysical applications to dynamical problems in machine learning. | [
"['Nayef Shkeir' 'Tobias Schäfer' 'Tobias Grafke']"
] |
null | null | 2406.13762 | null | null | http://arxiv.org/pdf/2406.13762v1 | 2024-06-19T18:22:32Z | 2024-06-19T18:22:32Z | Unveiling the Hidden Structure of Self-Attention via Kernel Principal
Component Analysis | The remarkable success of transformers in sequence modeling tasks, spanning various applications in natural language processing and computer vision, is attributed to the critical role of self-attention. Similar to the development of most deep learning models, the construction of these attention mechanisms rely on heuristics and experience. In our work, we derive self-attention from kernel principal component analysis (kernel PCA) and show that self-attention projects its query vectors onto the principal component axes of its key matrix in a feature space. We then formulate the exact formula for the value matrix in self-attention, theoretically and empirically demonstrating that this value matrix captures the eigenvectors of the Gram matrix of the key vectors in self-attention. Leveraging our kernel PCA framework, we propose Attention with Robust Principal Components (RPC-Attention), a novel class of robust attention that is resilient to data contamination. We empirically demonstrate the advantages of RPC-Attention over softmax attention on the ImageNet-1K object classification, WikiText-103 language modeling, and ADE20K image segmentation task. | [
"['Rachel S. Y. Teo' 'Tan M. Nguyen']"
] |
null | null | 2406.13768 | null | null | http://arxiv.org/pdf/2406.13768v1 | 2024-06-19T18:31:23Z | 2024-06-19T18:31:23Z | FastPersist: Accelerating Model Checkpointing in Deep Learning | Model checkpoints are critical Deep Learning (DL) artifacts that enable fault tolerance for training and downstream applications, such as inference. However, writing checkpoints to persistent storage, and other I/O aspects of DL training, are mostly ignored by compute-focused optimization efforts for faster training of rapidly growing models and datasets. Towards addressing this imbalance, we propose FastPersist to accelerate checkpoint creation in DL training. FastPersist combines three novel techniques: (i) NVMe optimizations for faster checkpoint writes to SSDs, (ii) efficient write parallelism using the available SSDs in training environments, and (iii) overlapping checkpointing with independent training computations. Our evaluation using real world dense and sparse DL models shows that FastPersist creates checkpoints in persistent storage up to 116x faster than baseline, and enables per-iteration checkpointing with negligible overhead. | [
"['Guanhua Wang' 'Olatunji Ruwase' 'Bing Xie' 'Yuxiong He']"
] |
null | null | 2406.13770 | null | null | http://arxiv.org/pdf/2406.13770v1 | 2024-06-19T18:38:11Z | 2024-06-19T18:38:11Z | Elliptical Attention | Pairwise dot-product self-attention is key to the success of transformers that achieve state-of-the-art performance across a variety of applications in language and vision. This dot-product self-attention computes attention weights among the input tokens using Euclidean distance, which makes the model prone to representation collapse and vulnerable to contaminated samples. In this paper, we propose using a Mahalanobis distance metric for computing the attention weights to stretch the underlying feature space in directions of high contextual relevance. In particular, we define a hyper-ellipsoidal neighborhood around each query to increase the attention weights of the tokens lying in the contextually important directions. We term this novel class of attention Elliptical Attention. Our Elliptical Attention provides two benefits: 1) reducing representation collapse and 2) enhancing the model's robustness as the Elliptical Attention pays more attention to contextually relevant information rather than focusing on some small subset of informative features. We empirically demonstrate the advantages of Elliptical Attention over the baseline dot-product attention and state-of-the-art attention methods on various practical tasks, including object classification, image segmentation, and language modeling across different data modalities. | [
"['Stefan K. Nielsen' 'Laziz U. Abdullaev' 'Rachel Teo' 'Tan M. Nguyen']"
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.