# AOR-Bench: Do Large Audio Language Models Over-Refuse Pseudo-Harmful Queries?

arXiv: https://arxiv.org/abs/2606.21147 · HTML: https://mlsanity.com/paper/2606.21147/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 19 of its 39 references that are in the map.
- First posted on arXiv 19 Jun 2026.

## 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, 19 of 19)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities](https://mlsanity.com/paper/2305.11000/) (2023)
- [Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models](https://mlsanity.com/paper/2402.17177/) (2024)
- [FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts](https://mlsanity.com/paper/2311.05608/) (2023)
- [Persona Vectors: Monitoring and Controlling Character Traits in Language Models](https://mlsanity.com/paper/2507.21509/) (2025)
- [CosyVoice 3: Towards In-the-wild Speech Generation via Scaling-up and Post-training](https://mlsanity.com/paper/2505.17589/) (2025)
- [A Comprehensive Survey of Hallucination in Large Language, Image, Video and Audio Foundation Models](https://mlsanity.com/paper/2405.09589/) (2024)
- [SCANS: Mitigating the Exaggerated Safety for LLMs via Safety-Conscious Activation Steering](https://mlsanity.com/paper/2408.11491/) (2024)
- [Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation](https://mlsanity.com/paper/2410.03415/) (2024)
- [MOSSBench: Is Your Multimodal Language Model Oversensitive to Safe Queries?](https://mlsanity.com/paper/2406.17806/) (2024)
- [Just Enough Shifts: Mitigating Over-Refusal in Aligned Language Models with Targeted Representation Fine-Tuning](https://mlsanity.com/paper/2507.04250/) (2025)
- [Personalized Safety in LLMs: A Benchmark and A Planning-Based Agent Approach](https://mlsanity.com/paper/2505.18882/) (2025)
- [When Large Language Models Meet Speech: A Survey on Integration Approaches](https://mlsanity.com/paper/2502.19548/) (2025)
- [SARSteer: Safeguarding Large Audio-Language Models via Safe-Ablated Refusal Steering](https://mlsanity.com/paper/2510.17633/) (2025)
- [Steering Multimodal Large Language Models Decoding for Context-Aware Safety](https://mlsanity.com/paper/2509.19212/) (2025)
- [EVOREFUSE: Evolutionary Prompt Optimization for Evaluation and Mitigation of LLM Over-Refusal to Pseudo-Malicious Instructions](https://mlsanity.com/paper/2505.23473/) (2025)
- [Investigating Safety Vulnerabilities of Large Audio-Language Models Under Speaker Emotional Variations](https://mlsanity.com/paper/2510.16893/) (2025)

## Abstract (arXiv)

Large Audio Language Models (LALMs) have demonstrated strong performance across a wide range of audio tasks. As they are increasingly deployed in real-world applications, ensuring their safety alignment has become more important. Although refusal mechanisms serve as a key safeguard by preventing LALMs from responding to harmful requests, they can also lead to over-refusal, where models incorrectly reject benign queries. This issue is especially challenging in the audio domain because speech that appears harmful in isolation may become benign when interpreted together with the surrounding acoustic context, such as background sounds. To study this problem, we introduce AOR-Bench (Audio Over-Refusal Benchmark), the first benchmark for over-refusal specifically designed for LALMs. AOR-Bench contains 3,000 pseudo-harmful audio samples across six scenario categories. Evaluating 12 representative LALMs from six major model families, we find that over-refusal is widespread (Figure 1) and uncover several important patterns in their safety judgments. As a preliminary effort to mitigate this issue, we further explore two lightweight strategies (e.g., Chain-of-Thought and activation steering) to reduce over-refusal.