File size: 9,160 Bytes
603f6e2 5ce778f 9388c5f 5ce778f cb85268 5ce778f cb85268 b249eea cb85268 b249eea cb85268 603f6e2 403af72 5ce778f 403af72 5ce778f 403af72 5ce778f 29e5c25 5ce778f 29e5c25 5ce778f 29e5c25 e2a3756 29e5c25 5ce778f 29e5c25 5ce778f 29e5c25 5ce778f 29e5c25 403af72 86100db 403af72 e2a3756 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 |
---
license: mit
task_categories:
- question-answering
language:
- en
tags:
- realtime
- news
configs:
- config_name: default
data_files:
- split: relive2025_01_15
path: 20250115_qa_public.jsonl
- split: relive2024_07_05
path: 20240705_qa_public.jsonl
- split: relive2024_06_29
path: 20240629_qa_public.jsonl
- split: relive2024_06_20
path: 20240620_qa_public.jsonl
- split: relive2024_06_13
path: 20240613_qa_public.jsonl
- split: relive2024_06_06
path: 20240606_qa_public.jsonl
- split: relive2024_05_25
path: 20240525_qa_public.jsonl
- split: relive2024_05_19
path: 20240519_qa_public.jsonl
- split: relive2024_05_18
path: 20240518_qa_public.jsonl
- split: rtqa2024_01_19
path: latest/20240119_qa_public.jsonl
- split: rtqa2024_01_12
path: past/2024/20240112_qa.jsonl
- split: rtqa2024_01_05
path: past/2024/20240105_qa.jsonl
- split: rtqa2023_12_22
path: past/2023/20231222_qa.jsonl
- split: rtqa2023_12_15
path: past/2023/20231215_qa.jsonl
- split: rtqa2023_12_08
path: past/2023/20231208_qa.jsonl
- split: rtqa2023_12_01
path: past/2023/20231201_qa.jsonl
- split: rtqa2023_11_24
path: past/2023/20231124_qa.jsonl
- split: rtqa2023_11_17
path: past/2023/20231117_qa.jsonl
- split: rtqa2023_11_10
path: past/2023/20231110_qa.jsonl
- split: rtqa2023_11_03
path: past/2023/20231103_qa.jsonl
- split: rtqa2023_10_27
path: past/2023/20231027_qa.jsonl
- split: rtqa2023_10_20
path: past/2023/20231020_qa.jsonl
- split: rtqa2023_10_13
path: past/2023/20231013_qa.jsonl
- split: rtqa2023_10_06
path: past/2023/20231006_qa.jsonl
- split: rtqa2023_09_29
path: past/2023/20230929_qa.jsonl
- split: rtqa2023_09_22
path: past/2023/20230922_qa.jsonl
- split: rtqa2023_09_15
path: past/2023/20230915_qa.jsonl
- split: rtqa2023_09_08
path: past/2023/20230908_qa.jsonl
- split: rtqa2023_09_01
path: past/2023/20230901_qa.jsonl
- split: rtqa2023_08_25
path: past/2023/20230825_qa.jsonl
- split: rtqa2023_08_18
path: past/2023/20230818_qa.jsonl
- split: rtqa2023_08_11
path: past/2023/20230811_qa.jsonl
- split: rtqa2023_08_04
path: past/2023/20230804_qa.jsonl
- split: rtqa2023_07_28
path: past/2023/20230728_qa.jsonl
- split: rtqa2023_07_21
path: past/2023/20230721_qa.jsonl
- split: rtqa2023_07_14
path: past/2023/20230714_qa.jsonl
- split: rtqa2023_07_07
path: past/2023/20230707_qa.jsonl
- split: rtqa2023_06_30
path: past/2023/20230630_qa.jsonl
- split: rtqa2023_06_23
path: past/2023/20230623_qa.jsonl
- split: rtqa2023_06_16
path: past/2023/20230616_qa.jsonl
- split: rtqa2023_06_09
path: past/2023/20230609_qa.jsonl
- split: rtqa2023_06_02
path: past/2023/20230602_qa.jsonl
- split: rtqa2023_05_26
path: past/2023/20230526_qa.jsonl
- split: rtqa2023_05_19
path: past/2023/20230519_qa.jsonl
- split: rtqa2023_05_12
path: past/2023/20230512_qa.jsonl
- split: rtqa2023_05_05
path: past/2023/20230505_qa.jsonl
- split: rtqa2023_04_28
path: past/2023/20230428_qa.jsonl
- split: rtqa2023_04_21
path: past/2023/20230421_qa.jsonl
- split: rtqa2023_04_14
path: past/2023/20230414_qa.jsonl
- split: rtqa2023_04_07
path: past/2023/20230407_qa.jsonl
- split: rtqa2023_03_31
path: past/2023/20230331_qa.jsonl
- split: rtqa2023_03_24
path: past/2023/20230324_qa.jsonl
- split: rtqa2023_03_17
path: past/2023/20230317_qa.jsonl
- split: rtqa2023_03_10
path: past/2023/20230310_qa.jsonl
- split: rtqa2023_03_03
path: past/2023/20230303_qa.jsonl
- split: rtqa2023_02_24
path: past/2023/20230224_qa.jsonl
- split: rtqa2023_02_17
path: past/2023/20230217_qa.jsonl
- split: rtqa2023_02_10
path: past/2023/20230210_qa.jsonl
- split: rtqa2023_02_03
path: past/2023/20230203_qa.jsonl
- split: rtqa2023_01_27
path: past/2023/20230127_qa.jsonl
- split: rtqa2023_01_20
path: past/2023/20230120_qa.jsonl
- split: rtqa2023_01_13
path: past/2023/20230113_qa.jsonl
- split: rtqa2023_01_06
path: past/2023/20230106_qa.jsonl
- split: rtqa2022_12_30
path: past/2022/20221230_qa.jsonl
- split: rtqa2022_12_23
path: past/2022/20221223_qa.jsonl
- split: rtqa2022_12_16
path: past/2022/20221216_qa.jsonl
- split: rtqa2022_12_09
path: past/2022/20221209_qa.jsonl
- split: rtqa2022_12_02
path: past/2022/20221202_qa.jsonl
- split: rtqa2022_11_25
path: past/2022/20221125_qa.jsonl
- split: rtqa2022_11_18
path: past/2022/20221118_qa.jsonl
- split: rtqa2022_11_11
path: past/2022/20221111_qa.jsonl
- split: rtqa2022_11_03
path: past/2022/20221104_qa.jsonl
- split: rtqa2022_10_28
path: past/2022/20221028_qa.jsonl
- split: rtqa2022_10_21b
path: past/2022/20221021_qa.jsonl
- split: rtqa2022_10_21a
path: past/2022/20221021_qa_public.jsonl
- split: rtqa2022_10_14b
path: past/2022/20221014_qa.jsonl
- split: rtqa2022_10_14a
path: past/2022/20220617-20221014_qa.jsonl
- split: rtqa2022_10_07
path: past/2022/20221007_qa.jsonl
- split: rtqa2022_09_30
path: past/2022/20220930_qa.jsonl
- split: rtqa2022_09_23
path: past/2022/20220923_qa.jsonl
- split: rtqa2022_09_16
path: past/2022/20220916_qa.jsonl
- split: rtqa2022_09_09
path: past/2022/20220909_qa.jsonl
- split: rtqa2022_09_02
path: past/2022/20220902_qa.jsonl
- split: rtqa2022_08_26b
path: past/2022/20220826_qa.jsonl
- split: rtqa2022_08_26a
path: past/2022/20220617-20220826_qa.jsonl
- split: rtqa2022_08_19
path: past/2022/20220819_qa.jsonl
- split: rtqa2022_08_12
path: past/2022/20220812_qa.jsonl
- split: rtqa2022_08_05
path: past/2022/20220805_qa.jsonl
- split: rtqa2022_07_29
path: past/2022/20220729_qa.jsonl
- split: rtqa2022_07_22b
path: past/2022/20220722_qa.jsonl
- split: rtqa2022_07_22a
path: past/2022/20220617-20220722_qa.jsonl
- split: rtqa2022_07_15b
path: past/2022/20220715_qa.jsonl
- split: rtqa2022_07_15a
path: past/2022/20220617-20220715_qa.jsonl
- split: rtqa2022_07_08b
path: past/2022/20220708_qa.jsonl
- split: rtqa2022_07_08a
path: past/2022/20220617-20220708_qa.jsonl
- split: rtqa2022_07_01
path: past/2022/20220701_qa.jsonl
- split: rtqa2022_06_24b
path: past/2022/20220624_qa.jsonl
- split: rtqa2022_06_24a
path: past/2022/20220617-20220624_qa.jsonl
- split: rtqa2022_06_17
path: past/2022/20220617_qa.jsonl
---
# relive-qa
Using RealtimeQA as a starting point for new articles + Q&A using a semi-automated format.
Also see:
- https://github.com/realtimeqa/realtimeqa_public (ended Jan 2024?)
- https://github.com/freshllms/freshqa (ongoing, last Dec 2024)
## Scraper process
Prerequisites: `pip install openai lxml cssselect requests xmltodict` and OpenAI API key
I've added these scripts:
- scrape.py : base script to load plain text from the latest WikiNews articles
- scrape_with_openai.py : pass scraped text to OpenAI's GPT-4o to generate questions and answers for each article
- scrape_morerecent_with_openai.py : scrape recent articles which WikiNews hasn't yet published from `Category:May 2024`, then use OpenAI's GPT-4o for Q&A
An LLM evaluated on this Q&A could read the WikiNews summary, the links collected from the story, and/or do its own web searches.
### Issues
- Prompt should discourage Q&A which are obvious, or stand out from alternative answers.
- Q&A should be based on new information in the article, and not general knowledge.
- Links and the article title could give away the answer as the subject of the article, rather than using reading comprehension.
- WikiNews articles may be niche / local stories, where facts are not known to an LLM unless it reads the specific article
## Original paper
Citation:
```
@inproceedings{
kasai2023realtime,
title={RealTime {QA}: What's the Answer Right Now?},
author={Jungo Kasai and Keisuke Sakaguchi and yoichi takahashi and Ronan Le Bras and Akari Asai and Xinyan Velocity Yu and Dragomir Radev and Noah A. Smith and Yejin Choi and Kentaro Inui},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
year={2023},
eprint={2207.13332},
url={https://openreview.net/forum?id=HfKOIPCvsv}
}
```
|