Datasets:

Modalities:
Image
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
pandas
License:
JasperHaozhe commited on
Commit
5a956a4
·
verified ·
1 Parent(s): b64d4c2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -3
README.md CHANGED
@@ -6,9 +6,19 @@ task_categories:
6
  - image-text-to-text
7
  pretty_name: PixelReasoner-RL-Data
8
  ---
 
 
9
 
10
- Under Construction!
 
 
 
 
 
 
11
 
12
- The RL data used in [Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learning](https://arxiv.org/abs/2505.15966). The code can be found at https://github.com/TIGER-AI-Lab/Pixel-Reasoner/tree/main
13
 
14
- Project page: https://tiger-ai-lab.github.io/Pixel-Reasoner/
 
 
 
6
  - image-text-to-text
7
  pretty_name: PixelReasoner-RL-Data
8
  ---
9
+ **Overview.**
10
+ The RL data for training [**Pixel Reasoner**: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learning](https://arxiv.org/abs/2505.15966).
11
 
12
+ **Details.**
13
+ The data includes 15,402 training queries with verifierable answers. The key fields include:
14
+ - question, answer, qid
15
+ - is_video: a flag to distinguish video and image queries
16
+ - image: a list of image paths.
17
+ - For video-based queries, the path list corresponds to 16 extracted frames from the original video
18
+ - For image-based queries, the path list corresponds to multiple images
19
 
20
+ **Note**: Remember to unzip `images.zip` and `videos.zip`, and replace the relative path in `image` field with the absolute path.
21
 
22
+ **Training Code**: The RL code can be found at https://github.com/TIGER-AI-Lab/Pixel-Reasoner/tree/main
23
+
24
+ **Project page**: https://tiger-ai-lab.github.io/Pixel-Reasoner/