aiqbench analysis
DeepWeaver addresses a critical bottleneck in retrieval-augmented generation by introducing Thought Block Chains to synthesize fragmented evidence into coherent, well-cited answers. This framework effectively prevents the collapse of diverse information into shallow summaries, offering a robust method for improving content sufficiency and citation grounding across multiple large language models. The accompanying LoQA benchmark provides a necessary high-density evaluation standard for measuring progress in open-ended question answering tasks.
DeepWeaver: Bridging the Evidence Synthesis Gap in Open-Ended Question Answering
arXiv:2608.18988v1 Announce Type: cross Abstract: Retrieve-then-generate pipelines are commonly used to produce deep-research answers for open-ended questions, but retrieval alone is insufficient: LLMs must organize noisy and fragmented evidence into comprehensive, well-cited answers. We refer to this process as evidence synthesis. However, direct generation often underuses evidence, misaligns citations, and collapses diverse information into shallow summaries, exposing an evidence synthesis gap between retrieval and generation. Thus, we propose DeepWeaver, a novel framework that weaves noisy retrieved evidence into comprehensive answers by maintaining Thought Block Chains (TBCs), a structured representation that groups claims, salient information, keywords, and supporting evidence. DeepWeaver uses subordinate TBCs to inspect residual evidence, commit TBC revisions, and discover new claims before final generation. We evaluate DeepWeaver on open-ended QA over both knowledge bases and the web, and introduce LoQA, a high-density benchmark for evidence synthesis. Across multiple LLMs, DeepWeaver improves content sufficiency, citation grounding, and detail preservation on LoQA, while achieving deeper insights and higher citation quality on DeepResearch Bench. These results show that evidence weaving is an effective mechanism for bridging retrieval and generation in open-ended QA. Our code is available at https://github.com/KlozeWang/DeepWeaver.
Abstract:Retrieve-then-generate pipelines are commonly used to produce deep-research answers for open-ended questions, but retrieval alone is insufficient: LLMs must organize noisy and fragmented evidence into comprehensive, well-cited answers. We refer to this process as evidence synthesis. However, direct generation often underuses evidence, misaligns citations, and collapses diverse information into shallow summaries, exposing an evidence synthesis gap between retrieval and generation. Thus, we propose DeepWeaver, a novel framework that weaves noisy retrieved evidence into comprehensive answers by maintaining Thought Block Chains (TBCs), a structured representation that groups claims, salient information, keywords, and supporting evidence. DeepWeaver uses subordinate TBCs to inspect residual evidence, commit TBC revisions, and discover new claims before final generation. We evaluate DeepWeaver on open-ended QA over both knowledge bases and the web, and introduce LoQA, a high-density benchmark for evidence synthesis. Across multiple LLMs, DeepWeaver improves content sufficiency, citation grounding, and detail preservation on LoQA, while achieving deeper insights and higher citation quality on DeepResearch Bench. These results show that evidence weaving is an effective mechanism for bridging retrieval and generation in open-ended QA. Our code is available at this https URL.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2608.18988 [cs.CL]
(or arXiv:2608.18988v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2608.18988 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 282 words
Related coverage
Enabling a new model for healthcare with AI co-clinician
Google DeepMind · Apr 30, 2026
State of the Field: AI for Epistemics and Coordination
LessWrong · Oct 2, 2026
Links #3: 2026/06 Part 1
LessWrong · Jun 15, 2026
LakeQuest: A Three-Domain Benchmark for Grounded Question Answering across Data Lakes
arXiv · Jul 23, 2026