CHRONOS addresses the challenge of open-domain Timeline Summarization (TLS) by iteratively generating chronological summaries of news events. It is designed for researchers and practitioners in Natural Language Processing and Information Retrieval interested in advanced summarization techniques. The project offers a novel retrieval-based approach that improves efficiency and scalability for TLS tasks.

How It Works

CHRONOS employs an iterative self-questioning strategy. The model poses questions about a given topic and retrieved documents, then uses the answers to refine its understanding and generate a chronological summary. This approach allows for dynamic exploration of information and adaptation to new events, outperforming static methods.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Requires API keys for Qwen/GPT (DashScope/OpenAI), Bing Web Search, and optionally Jina.
• Example usage: python main.py --model_name "$model" --max_round "$round" --dataset open --output "$output_dir" --question_exs
• Official Demo: https://modelscope.cn/studios/vickywu1022/CHRONOS
• Paper: https://arxiv.org/abs/2501.00888

Highlighted Details

• Introduces the Open-TLS dataset, a large-scale dataset for open-domain TLS.
• Achieves comparable results to state-of-the-art closed-domain TLS methods.
• Demonstrates significant improvements in efficiency and scalability.
• Supports iterative self-questioning for dynamic information retrieval.

Maintenance & Community

The project is associated with Alibaba-NLP and the NAACL 2025 paper "Unfolding the Headline: Iterative Self-Questioning for News Retrieval and Timeline Summarization." Further community or maintenance details are not explicitly provided in the README.

Licensing & Compatibility

The README does not specify a license. Users should verify licensing for API key usage (Qwen/GPT, Bing, Jina) and for any commercial or closed-source integration.

Limitations & Caveats

The project requires multiple third-party API keys for full functionality, which may incur costs. The setup involves replacing placeholder API keys in source files. The project is presented as code for a specific research paper, and its long-term maintenance status is not detailed.