Researchers have developed TimelyRAG, a new framework designed to improve question-answering over documents that are frequently updated. Unlike existing methods that treat document versions as separate, TimelyRAG accounts for overlapping and evolving document structures, such as legal amendments. The framework integrates temporal distance into the retrieval process to ensure queries are matched with the most relevant document versions. To evaluate its effectiveness, a new benchmark called TimelyQABench was created, focusing on regulation-heavy domains. Experiments demonstrated significant improvements in retrieval accuracy, with gains up to 28.6% in nDCG@10. AI
IMPACT Enhances reliability of QA systems in dynamic information environments, crucial for legal and policy domains.
RANK_REASON The cluster contains a research paper detailing a new framework and benchmark for question answering over evolving documents. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
- alphaXiv
- arXiv
- CatalyzeX
- DagsHub
- Gotit.pub
- Hugging Face
- retrieval-augmented generation
- ScienceCast
- TimelyQABench
- TimelyRAG
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
