# HarmMetric Eval: Benchmarking Metrics and Judges for LLM Harmfulness Assessment

arXiv: https://arxiv.org/abs/2509.24384 · HTML: https://mlsanity.com/paper/2509.24384/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 38 of its 77 references that are in the map.
- First posted on arXiv 29 Sep 2025.

## Placed in

- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 30 of 38)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Jailbreaking Black Box Large Language Models in Twenty Queries](https://mlsanity.com/paper/2310.08419/) (2023)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models](https://mlsanity.com/paper/2310.04451/) (2023)
- [Large Language Models](https://mlsanity.com/paper/2307.05782/) (2023)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [MTEB: Massive Text Embedding Benchmark](https://mlsanity.com/paper/2210.07316/) (2022)
- [How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs](https://mlsanity.com/paper/2401.06373/) (2024)
- [GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts](https://mlsanity.com/paper/2309.10253/) (2023)
- [JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models](https://mlsanity.com/paper/2404.01318/) (2024)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- [NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models](https://mlsanity.com/paper/2405.17428/) (2024)
- [Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation](https://mlsanity.com/paper/2310.06987/) (2023)
- [Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations](https://mlsanity.com/paper/2310.06387/) (2023)
- [Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment](https://mlsanity.com/paper/2308.05374/) (2023)
- [Jailbreak Attacks and Defenses Against Large Language Models: A Survey](https://mlsanity.com/paper/2407.04295/) (2024)
- [Scalable and Transferable Black-Box Jailbreaks for Language Models via Persona Modulation](https://mlsanity.com/paper/2311.03348/) (2023)

## Built on by (most cited first, 15 of 15)

- [TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking](https://mlsanity.com/paper/2605.30883/) (2026)
- [DARWIN: Evolving Jailbreak Adversary and Guardrail for LLM Safety Evaluation and Protection](https://mlsanity.com/paper/2607.19829/) (2026)
- [NonTextual Target Attack](https://mlsanity.com/paper/2510.02999/) (2025)
- [Towards Identification and Intervention of Safety-Critical Parameters in Large Language Models](https://mlsanity.com/paper/2604.08297/) (2026)
- [When Behavioral Safety Evaluation Fails: A Representation-Level Perspective](https://mlsanity.com/paper/2606.08044/) (2026)
- [Quantifying LLM Safety Degradation Under Repeated Attacks Using Survival Analysis](https://mlsanity.com/paper/2605.12869/) (2026)
- [AICompanionBench: Benchmarking LLMs-as-Judges for AI Companion Safety](https://mlsanity.com/paper/2606.04867/) (2026)
- [Distillability of LLM Security Logic: Predicting Attack Success Rate of Outline Filling Attack via Ranking Regression](https://mlsanity.com/paper/2511.22044/) (2025)
- [From Performance to Purpose: A Sociotechnical Taxonomy for Evaluating Large Language Model Utility](https://mlsanity.com/paper/2602.20513/) (2026)
- [Where Does Toxicity Live? Mechanistic Localization and Targeted Suppression in Language Models](https://mlsanity.com/paper/2605.27997/) (2026)
- [PHANTOM: A Large-Scale Dataset of Multimodal Adversarial Attacks for Vision-Language Models](https://mlsanity.com/paper/2606.24388/) (2026)
- [CLEAR: Continuous Latent Adapter Routing for Utility-Preserving LLM Safety Alignment](https://mlsanity.com/paper/2608.21278/) (2026)
- [Who Judges the Judges? A Chinese Safety QA Benchmark for Evaluating LLM Responses and Safety Judges](https://mlsanity.com/paper/2609.01210/) (2026)
- [ASCII Attack: Recontextualising Harmful Requests as Artistic Critique in Large Language Models](https://mlsanity.com/paper/2609.02215/) (2026)
- [Red-TTT: Test-Time Training for Automated Jailbreaking Large Language Models](https://mlsanity.com/paper/2610.05282/) (2026)

## Abstract (arXiv)

The potential of large language models (LLMs) to generate harmful content poses a significant safety risk for data management, as LLMs are increasingly being used as engines for data generation. To assess this risk, numerous harmfulness evaluation metrics and judges have been proposed. However, due to differences in their formats and scales, these metrics may yield inconsistent evaluation results on LLM-generated harmful data, undermining their credibility in practice. To address this gap, we present HarmMetric Eval, a systematic benchmark for assessing the quality of harmfulness metrics and judges with varying formats and scales. HarmMetric Eval includes a high-quality dataset comprising representative harmful prompts paired with harmful and non-harmful LLM outputs across multiple fine-grained categories, along with a unified scoring mechanism to reward the metrics for correctly ranking harmful outputs over non-harmful ones. Extensive experiments on HarmMetric Eval yield a surprising finding: conventional reference-based metrics such as ROUGE and METEOR can outperform LLM-based judges in fine-grained harmfulness evaluation, challenging prevailing assumptions about LLMs' superiority in this domain. To reveal the reasons behind this finding, we provide a fine-grained analysis to explain the limitations of LLM-based judges on rating irrelevant or useless LLM outputs. Motivated by these insights, we design an improved harmfulness judge that explicitly incorporates fine-grained harmfulness criteria in its prompt template and leverages reference-based metrics for lightweight fine-tuning of its base LLM. The resulting judge achieves state-of-the-art evaluation effectiveness on HarmMetric Eval.