No AI summary available for this article.
Why It Matters
Factual question answering (QA) typically assumes a single canonical answer, obscuring whether large language models (LLMs) retain divergent accounts of long-tail facts.
Provenance
Discovered via ArXiv and published by ArXiv.
Key Claims
Original description
Factual question answering (QA) typically assumes a single canonical answer, obscuring whether large language models (LLMs) retain divergent accounts of long-tail facts. To address this gap, we introduce ElephantBench, a closed-book knowledge probe comprising 1,094 questions generated through an auditable graph-based pipeline. The pipeline retrieves related documents from a low-exposure web corpus, identifies naturally occurring disagreements, and converts them into multi-account QA records. Each answer is verified against the originating documents and authoritative public web sources and is t...
Discovered via ArXiv
Research papers and preprints from arXiv.
Publisher: arxiv.org
ID: http://arxiv.org/abs/2608.28478v1 · Indexed about 5 hours ago