AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

AudioTrust: Benchmarking The Multifaceted Trustworthiness of Audio Large Language Models

Kai Li, Can Shen, Yile Liu, Jirui Han, Kelong zheng, Xuechao Zou, Lionel Z. WANG, Shun Zhang, Xingjian Du, Hanjun Luo, Yingbin Jin, Xinxin Xing and 22 more

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Speech and audio language models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add privacy-enhancing prompts (ablation)ALLMs (average) · AudioTrust privacy, direct leakagerefusal rate: 29.99 -> 63.77 (GPT-4o evaluator)Improved
add privacy-enhancing prompts (ablation)ALLMs (average) · AudioTrust privacy, inference leakagerefusal rate: 9.02 -> 12.12No clear effect
add privacy-aware prompt to Qwen2.5-Omni (ablation)Qwen2.5-Omni · AudioTrust privacy, direct leakagerefusal rate: 0.00 -> 1.17No clear effect
closed-source vs open-source modelsALLMs · AudioTrust robustnessscore: closed average generally above open-sourceImproved
closed-source vs open-source models on voice cloning defenseALLMs · AudioTrust authentication, VCSdefense success: open avg 45.0 vs closed avg 44.9Mixed
closed-source vs open-source on identity verification bypassALLMs · AudioTrust authentication, IVBdefense success: open avg 55.3 vs closed 97.2Improved
stricter speech-text prompting against voice cloning (ablation)ALLMs · AudioTrust authenticationdefense successImproved
closed-source vs open-source on fairnessALLMs · AudioTrust fairnessGamma: GPT-4o Audio 0.926 vs Gemini-2.5 Pro 0.319 (stereotype)Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy