Unfolding the Headline: Iterative Self-Questioning for News Retrieval and Timeline Summarization

2025-01-01

AI Summary

Open-domain timeline summarization (TLS) faces severe challenges under information overload—automatically constructing coherent, causally grounded timelines from massive, heterogeneous news streams. Method: We propose CHRONOS, the first framework leveraging causal reasoning–driven iterative self-questioning to guide LLMs in dynamically retrieving reports, modeling event causality, and generating timeline graphs. It integrates iterative retrieval-augmented generation (RAG), dynamic query generation, and structured event graph construction. Contributions/Results: (1) We introduce Open-TLS, the first open-domain TLS benchmark annotated by professional journalists; (2) CHRONOS significantly outperforms existing baselines on open-domain TLS, matching state-of-the-art performance in closed-domain settings; (3) Extensive ablation and case studies empirically validate the effectiveness and robustness of the self-questioning paradigm for real-world news temporal understanding.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models
• Reasoning under Uncertainty: Causality

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

In the fast-changing realm of information, the capacity to construct coherent timelines from extensive event-related content has become increasingly significant and challenging. The complexity arises in aggregating related documents to build a meaningful event graph around a central topic. This paper proposes CHRONOS - Causal Headline Retrieval for Open-domain News Timeline SummarizatiOn via Iterative Self-Questioning, which offers a fresh perspective on the integration of Large Language Models (LLMs) to tackle the task of Timeline Summarization (TLS). By iteratively reflecting on how events are linked and posing new questions regarding a specific news topic to gather information online or from an offline knowledge base, LLMs produce and refresh chronological summaries based on documents retrieved in each round. Furthermore, we curate Open-TLS, a novel dataset of timelines on recent news topics authored by professional journalists to evaluate open-domain TLS where information overload makes it impossible to find comprehensive relevant documents from the web. Our experiments indicate that CHRONOS is not only adept at open-domain timeline summarization, but it also rivals the performance of existing state-of-the-art systems designed for closed-domain applications, where a related news corpus is provided for summarization.

Problem

• Information Retrieval
• Timeline Generation
• Event Sequencing

Innovation

• CHRONOS
• Open-TLS
• Temporal Narrative Construction