Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/Alibaba-NLP/CHRONOS?tab=readme-ov-file
Repo for Paper "Unfolding the Headline: Iterative Self-Questioning for News Retrieval and Timeline Summarization"
https://github.com/Alibaba-NLP/CHRONOS?tab=readme-ov-file
Last synced: 16 days ago
JSON representation
Repo for Paper "Unfolding the Headline: Iterative Self-Questioning for News Retrieval and Timeline Summarization"
- Host: GitHub
- URL: https://github.com/Alibaba-NLP/CHRONOS?tab=readme-ov-file
- Owner: Alibaba-NLP
- Created: 2024-12-31T11:35:15.000Z (about 1 month ago)
- Default Branch: main
- Last Pushed: 2025-01-22T10:42:48.000Z (18 days ago)
- Last Synced: 2025-01-22T11:31:45.610Z (18 days ago)
- Language: Python
- Size: 15.8 MB
- Stars: 118
- Watchers: 8
- Forks: 10
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- Changelog: news_keywords.py
Awesome Lists containing this project
- awesome-llm-agent - CHRONOS
README
# CHRONOS: News Timeline Summarization
[](https://pytorch.org/)[](https://arxiv.org/abs/)
📑 **Paper:** https://arxiv.org/abs/2501.00888
🌏 **Chinese Web Demo:** https://modelscope.cn/studios/vickywu1022/CHRONOS
## 🚀Overview
- We propose CHRONOS, a novel **retrieval-based** approach to Timeline Summarization (TLS) by **iteratively posing questions** about the topic and the retrieved documents to generate chronological summaries.
- We construct an **up-to-date dataset for open- domain TLS**, which surpasses existing public datasets in terms of both size and the duration of timelines.
- Experiments demonstrate that our method is effective on open-domain TLS and achieves comparable results with state-of-the-art methods of closed-domain TLS, with significant improvements in **efficiency and scalability**.
## ⚗️ OPEN-TLS Dataset
We release our Open-TLS dataset for open-domain Timeline Summarization.
The target news query is presented in `news_keywords.py` and the ground truth timeline is presented in `data/open/{NEWS_KEYWORD}/timelines.jsonl` following the below format:
```json
[["YYY-MM-DDT00:00:00", ["", "", ""]]]
```
Statistics of Open-TLS are:
## 🛠 Running CHRONOS
### Step 1. Dependencies
```bash
pip install -r requirements.txt
```
### Step 2. Exampled Questions Generation
The second step is to construct a topic-questions example pool for datasets in `data/`.
```python
python question_exampler.py
```
Or, you can use our provided `data/question_examples.json`, which contains examples for the *Crisis, T17 and Open-TLS* datasets.
### Step 3. Running CHRONOS
We have released the code of CHRONOS to complete open-domain Timeline Summarization task. You may also refer to our [modelscope repo](https://modelscope.cn/studios/vickywu1022/CHRONOS/file/view/master?fileName=app.py&status=1) to build an app with `streamlit`.
#### Replacing Keys
Before running, please replace the placeholder with your own API keys in `src/model.py` to call either Qwen or GPT models.
```python
DASHSCOPE_API_KEY = "YOUR_API_KEY"
OPENAI_API_KEY = "YOUR_API_KEY"
```
Please also replace it with your own BING Web Search API key in `src/searcher.py` to search news from the Internet.
```python
BING_SEARCH_KEY = "YOUR_API_KEY"
```
If you want the CHRONOS to use the full page instead of only the snippet, please replace your own JINA key in `src/reader.py`.
```python
JINA_API_KEY = "YOUR_API_KEY"
```
#### Running Script
To experiment with the Open-TLS dataset, run:
```python
python main.py \
--model_name "$model" \
--max_round "$round" \
--dataset open \
--output "$output_dir" \
--question_exs
```
where `"$round"` is the maximum self-questioning round and `"$output_dir"` sets the output directory containing: (1) retrieved news, (2) generated timelines and (3) evaluation scores.
## 📝 Citation
```bigquery
@article{wu2025unfoldingheadlineiterativeselfquestioning,
title={Unfolding the Headline: Iterative Self-Questioning for News Retrieval and Timeline Summarization},
author={Weiqi Wu and Shen Huang and Yong Jiang and Pengjun Xie and Fei Huang and Hai Zhao},
year={2025},
eprint={2501.00888},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2501.00888},
}
```
## Star History
[](https://star-history.com/#Alibaba-NLP/CHRONOS&Date)