SPIN Unprocessed October 23, 2026 ai_technology research
Behind Aggregate Scores: What Systematically Varying Arena-Hard's LLM Judge Reveals
View original on arxiv.orgOverview
arXiv:2610.30512v1 Announce Type: new Abstract: LLM judges are used to determine which of two answers is better. Trusting these verdicts requires knowing how repeatable they are and how much they depend on the judge's instructions. We show that similar aggregate scores can conceal substantial changes in individual verdicts, both when the same setup is repeated and when the instructions change. Using 500 Arena-Hard v2.0 prompts with two models' answers held fixed, we run GPT-4.1 as the judge unde
SpinGraph analysis pending — check back after processing.
Ask AI about this story
Opens with the SpinGraph .md URL and structured context — one click, prompt included.
More from arXiv Computation and Language
View all →- WikiOriginQA: Automated Construction of a QA Dataset with Knowledge Origins for Analyzing Cultural Bias
- Spectral Smoothing and Spiking Perturbations Reconfigure Capability Profiles and Selective Task Performance in Large Language Models
- Compiling Autoregressive LLM Control into Policy-Exact Parallel Inference for Hierarchical RAG
- What the Specification Did Not Say: Specification-Oracle Gaps in Test-Graded Coding-Agent Benchmarks
- IFC-MCQ: Automatically Generating Multiple-Choice Benchmarks for Evaluating LLM Agents on IFC Data Retrieval
- Protegrity-PII Technical Report
Markdown (.md) · JSON-LD schema (.json) · Machine-readable for AI & GEO