RESEARCHMonitorWATCHLIST
How Reproducible Are Evaluation Conclusions? A Self-Audit of LLM-Inferred Prompt Structure
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
ArXiv paper finds LLM evaluation conclusions are unstable across prompt variants, undermining confidence in ranked model benchmark tables.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 25 September 2026
- Collected by OneBench
- 26 Sept 2026, 03:01 UK
Stored source excerpt
arXiv:2609.30074v1 Announce Type: new Abstract: Evaluations of LLM systems routinely average over small prompt sets and report models as a ranked table. We ask how…
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.