# SAFETY-J: Evaluating Safety with Critique

arXiv: https://arxiv.org/abs/2407.17075 · HTML: https://mlsanity.com/paper/2407.17075/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 24 Jul 2024.

## Builds on (most cited first, 20 of 20)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset](https://mlsanity.com/paper/2307.04657/) (2023)
- [WildChat: 1M ChatGPT Interaction Logs in the Wild](https://mlsanity.com/paper/2405.01470/) (2024)
- [Recipes for building an open-domain chatbot](https://mlsanity.com/paper/2004.13637/) (2020)
- [GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts](https://mlsanity.com/paper/2309.10253/) (2023)
- [Artificial Intelligence, Values and Alignment](https://mlsanity.com/paper/2001.09768/) (2020)
- [Self-critiquing models for assisting human evaluators](https://mlsanity.com/paper/2206.05802/) (2022)
- [A Holistic Approach to Undesired Content Detection in the Real World](https://mlsanity.com/paper/2208.03274/) (2022)
- [PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization](https://mlsanity.com/paper/2306.05087/) (2023)
- [A New Generation of Perspective API: Efficient Multilingual Character-level Transformers](https://mlsanity.com/paper/2202.11176/) (2022)
- [Shepherd: A Critic for Language Model Generation](https://mlsanity.com/paper/2308.04592/) (2023)
- [CValues: Measuring the Values of Chinese Large Language Models from Safety to Responsibility](https://mlsanity.com/paper/2307.09705/) (2023)
- [On the Safety of Conversational Models: Taxonomy, Dataset, and Benchmark](https://mlsanity.com/paper/2110.08466/) (2021)
- [ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectors](https://mlsanity.com/paper/2402.16444/) (2024)
- [Align on the Fly: Adapting Chatbot Behavior to Established Norms](https://mlsanity.com/paper/2312.15907/) (2023)
- [LLMCRIT: Teaching Large Language Models to Use Criteria](https://mlsanity.com/paper/2403.01069/) (2024)

## Built on by (most cited first, 3 of 3)

- [O1 Replication Journey -- Part 2: Surpassing O1-preview through Simple Distillation, Big Progress or Bitter Lesson?](https://mlsanity.com/paper/2411.16489/) (2024)
- [Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks](https://mlsanity.com/paper/2504.18838/) (2025)
- [How AI Assistants Respond to Repeated Abuse](https://mlsanity.com/paper/2609.17547/) (2026)

## Abstract (arXiv)

The deployment of Large Language Models (LLMs) in content generation raises significant safety concerns, particularly regarding the transparency and interpretability of content evaluations. Current methods, primarily focused on binary safety classifications, lack mechanisms for detailed critique, limiting their utility for model improvement and user trust. To address these limitations, we introduce SAFETY-J, a bilingual generative safety evaluator for English and Chinese with critique-based judgment. SAFETY-J utilizes a robust training dataset that includes diverse dialogues and augmented query-response pairs to assess safety across various scenarios comprehensively. We establish an automated meta-evaluation benchmark that objectively assesses the quality of critiques with minimal human intervention, facilitating scalable and continuous improvement. Additionally, SAFETY-J employs an iterative preference learning technique to dynamically refine safety assessments based on meta-evaluations and critiques. Our evaluations demonstrate that SAFETY-J provides more nuanced and accurate safety evaluations, thereby enhancing both critique quality and predictive reliability in complex content scenarios. To facilitate further research and application, we open-source SAFETY-J's training protocols, datasets, and code at https://github.com/GAIR-NLP/Safety-J.