# LLM Judges Inconsistently Disagree Across Safety Criteria and Harm Categories

arXiv: https://arxiv.org/abs/2605.31381 · HTML: https://mlsanity.com/paper/2605.31381/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 37 references that are in the map.
- First posted on arXiv 29 May 2026.

## Builds on (most cited first, 18 of 18)

- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods](https://mlsanity.com/paper/2412.05579/) (2024)
- [Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models](https://mlsanity.com/paper/2404.18796/) (2024)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2406.07791/) (2024)
- [LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks](https://mlsanity.com/paper/2406.18403/) (2024)
- [Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judges](https://mlsanity.com/paper/2406.12624/) (2024)
- [The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm](https://mlsanity.com/paper/2406.18682/) (2024)
- [From Hard Refusals to Safe-Completions: Toward Output-Centric Safety Training](https://mlsanity.com/paper/2508.09224/) (2025)
- [Neither Valid nor Reliable? Investigating the Use of LLMs as Judges](https://mlsanity.com/paper/2508.18076/) (2025)
- [Rating Roulette: Self-Inconsistency in LLM-As-A-Judge Frameworks](https://mlsanity.com/paper/2510.27106/) (2025)
- [How Reliable is Multilingual LLM-as-a-Judge?](https://mlsanity.com/paper/2505.12201/) (2025)
- [Safer or Luckier? LLMs as Safety Evaluators Are Not Robust to Artifacts](https://mlsanity.com/paper/2503.09347/) (2025)
- [Noisy but Valid: Robust Statistical Evaluation of LLMs with Imperfect Judges](https://mlsanity.com/paper/2601.20913/) (2026)
- [Dialectal Toxicity Detection: Evaluating LLM-as-a-Judge Consistency Across Language Varieties](https://mlsanity.com/paper/2411.10954/) (2024)
- [Judge Reliability Harness: Stress Testing the Reliability of LLM Judges](https://mlsanity.com/paper/2603.05399/) (2026)
- [Skewed Score: A statistical framework to assess autograders](https://mlsanity.com/paper/2507.03772/) (2025)
- [The Judge Who Never Admits: Hidden Shortcuts in LLM-based Evaluation](https://mlsanity.com/paper/2602.07996/) (2026)

## Built on by (most cited first, 1 of 1)

- [Manac'a-1B-Instruct-2: Reproducible Safety Alignment for a Native Brazilian-Portuguese LLM. Negative DPO Results, a Controlled Composition Scan, and a Silent Failure Mode of the LLM Judge](https://mlsanity.com/paper/2610.20570/) (2026)

## Abstract (arXiv)

We evaluate the consistency of automated judges in conducting a multi-dimensional safety evaluation in a reference-free setup. Our results indicate that Large Language Models are unreliable judges in identifying safety issues related to machine-generated advice in regulated domains such as finance, although they are more reliable at identifying more overt forms of unsafe/harmful content such as violence. The degree of inconsistency in a model's judgments can vary significantly by the chosen safety criteria and can be impacted by the language of the content and its linguistic style as well. Finally, there is high disagreement among different judges for the same output, across domains, safety criteria, and languages. These findings provide new insights on the practice of using LLMs as evaluators and offer several recommendations for practitioners on how to use automated judges in practical scenarios.