Reproduce “TimelineAgent: An LLM-based Framework for Query-Centric Event Timeline Extraction Across Multiple Documents” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Qingkai Min, Yunxin Bao, Yifeng Lu, Zhenyu Feng, Yulong Chen, Andreas Vlachos, Yue Zhang
TimelineAgent builds query-focused timelines from news reports by selecting relevant events, inferring normalized dates, filtering ineligible events, and merging mentions of the same event across documents. It routes reusable feature skills through successive reasoning steps for entity, aspect, and seed-event queries. On QUEST, it outperforms reported baselines overall; however, Qwen3-32B large SFT scores slightly higher on aspect queries.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.