RESEARCHInvestigateNEXT 12 MONTHS
Relevance Is Not Sufficient Evidence: Detecting Evidence Gaps Before Generation in RAG
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Research shows 12 LLMs fail to abstain on insufficient-evidence RAG questions, answering 40.0-99.3% of queries despite missing facts.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 30 September 2026
- Collected by OneBench
- 1 Oct 2026, 03:01 UK
Stored source excerpt
arXiv:2609.37469v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) grounds large language models in external sources, but retrieved passages often name the right entities without providing…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.