Mediocrity is the key for LLM as a Judge Anchor Selection

arXiv:2603.16848 · cs.CL · Submitted 2026-03-17 · Read on arXiv

cs.CL

Submitted: 2026-03-17

Updated: 2026-09-02

Code: https://github.com/IBM/Anchor-Selection

Terminology

• Benchmark — 20× in this paper · explained in 2 episodes
• Alignment — 7× in this paper · explained in 1 episode
• LLM-as-a-Judge — 5× in this paper · explained in 3 episodes
• Pairwise Comparisons — 5× in this paper · explained in 2 episodes
• Aggregation — 4× in this paper · explained in 1 episode
• Benchmarking — 4× in this paper · explained in 3 episodes
• LLM judge — 3× in this paper · explained in 2 episodes
• Robustness — 3× in this paper · explained in 5 episodes
• Accuracy — 2× in this paper · explained in 1 episode
• Reinforcement Learning — 2× in this paper · explained in 4 episodes
• Sample Efficiency — 2× in this paper · explained in 4 episodes
• alpha — 2× in this paper · explained in 1 episode
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes

Sources

• Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• The Falcon Series of Open Language Models
• Mixtral of Experts
• Naturally Occurring Feedback is Common, Extractable and Useful
• Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators
• From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline
• A Survey on LLM-as-a-Judge
• gpt-oss-120b & gpt-oss-20b Model Card
• Beyond Pairwise: Empowering LLM Alignment With Ranked Choice Modeling
• Gemma 3 Technical Report
• Zero-shot Benchmarking: A Framework for Flexible and Scalable Automatic Evaluation of Language Models
• TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them
• Verbosity Bias in Preference Labeling by Large Language Models
• Qwen3 Technical Report
• Qwen2 Technical Report
• Qwen2.5 Technical Report
• Yi: Open Foundation Models by 01.AI

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering