arXiv：cs.AI· Mohammad Khalooei, Mohammad Sabokrou·· 13 小时前
位置是否重要？黑盒因果审计揭示 LALM 幻觉中的模型依赖音频 grounding
Does Location Matter? A Black-Box Causal Audit Reveals Model-Dependent Audio Grounding in LALM Hallucination
AI 导读
研究提出黑盒因果测试，通过 CLAP 相似度定位、频谱门控移除并对照随机窗口，在三个 LALM 上验证音频 grounding。在 Qwen2-Audio-7B 上，固定时长后局部移除在每个网格单元均优于随机移除，而原始检测值仍因模型而异。结合 blank 与语言先验对比的门控集成对比解码，将 Qwen2-Audio 准确率从 57.1% 提升至 71.9%。
正文
Abstract:Large audio-language models (LALMs) frequently exhibit object hallucination, affirming absent sounds. The dominant mitigation, Audio-Aware Decoding (AAD), contrasts logits under real audio against an all-zero "blank" waveform, flagged by prior review as out-of-distribution. We introduce a black-box causal test that isolates the claimed evidence itself: localize a claim's evidence region via CLAP similarity, remove it by spectral gating, and compare against a random-window control across a 2x2x2 grid on three LALMs. On Qwen2-Audio-7B, localized ablation's raw AUROC trails blank's (0.613 vs. 0.721); this is a duration confound, not a localization failure, since blank removes the entire clip versus the grid's 2-4 s. Holding duration fixed, localization beats random removal at every cell on both models. Raw detection value remains model-dependent, especially on Audio Flamingo 3. The same evidence corrects decoding: contrastive decoding combining blank and language-prior contrasts, gated and ensembled, raises Qwen2-Audio accuracy from 57.1% to 71.9%, beating AAD alone. A unified causal-grounding profile does not transfer between models, indicating different grounding regimes. The source code and a detailed report of our findings, including identified limitations, are available at this https URL.
| Comments: | 5 pages, 1 figures, 4 tables. Submitted to IEEE ICASSP 2027 |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Sound (cs.SD); Audio and Speech Processing (eess.AS) |
| Cite as: | arXiv:2610.13112 [cs.AI] |
| (or arXiv:2610.13112v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.13112 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Mohammad Khalooei [view email]
[v1]
Fri, 9 Oct 2026 07:06:24 UTC (37 KB)
来源：arXiv：cs.AI · arxiv.org