CORTEXA
← Browse
arxivcs.SDcs.AI2026-06-30

Adaptive Perturbation Selection for Contrastive Audio Decoding

Aaron Isidore Grace, Zhouyuan Huo, Weiran Wang

Large audio-language models (LALMs) frequently hallucinate by overriding acoustic evidence with language priors. While contrastive decoding (CD) offers training-free mitigation, existing methods rely on blunt perturbations like masking or noise, leaving structured audio transformations unexplored. We explore this design space by evaluating a diverse library of targeted audio perturbations and adaptively selecting the optimal negative branch for each task and example. First, we improve upon earlier prompt engineering by showing that a simple binary yes/no constraint reduces the model's tendency to falsely confirm absent audio features. Second, evaluating our library across temporal, spectral, frequency, and amplitude domains reveals that optimal transformations are highly task-dependent; for instance, reversing the audio array disrupts temporal coherence, raising accuracy on the temporal order task from 74.7% to 81.4%. Finally, we trained a light-weight perturbation selector on model hidden states to dynamically route negative branches, yielding an additional +4.3% gain on the existence task.

View free PDFSource page

Related papers

arxivcs.SDcs.AI2026-07-22

Audio-Zero: Label-Free Self-Evolution for Fine-Grained Audio Reasoning

Siqian Tong, Xuan Li, Chaozhuo Li, Baolong Bi, Yiwei Wang, Yujun Cai, et al.

Large Audio Language models (LALMs) have made rapid progress on acoustic understanding, yet they still struggle with fine-grained audio reasoning (e.g., recognizing event order, repetitions and duration). Existing post-training methods heavily rely on expensive external labels or…

View free PDFSource page
arxivcs.SDcs.AIeess.AS2026-06-27

ALM2Vec: Learning Audio Embeddings for Universal Audio Retrieval with Large Audio-Language Models

Fengjie Lu, Chenang Jiang, Jiarui Hai, Helin Wang, Aaron Yee

Recent advances in language--audio retrieval have been largely driven by contrastive dual-encoder architectures that align audio and text in a shared embedding space. While effective, existing retrieval embeddings are primarily optimized for audio--caption matching, limiting thei…

View free PDFSource page
arxivcs.SDcs.AI2026-07-13

Encoder-Side Neuron Identification and Amplification for Acoustic Perception in Large Audio-Language Models

Yu-Han Huang, Chih-Kai Yang, Ke-Han Lu, An-Yu Cheng, Hung-yi Lee

Large audio-language models (LALMs) often underperform on fine-grained, non-semantic attributes of speech, such as a speaker's emotion, despite strong performance on speech content. Improving this without the cost of retraining calls for an effective inference-time intervention,…

View free PDFSource page
arxivcs.CLcs.AIcs.LGcs.SDeess.AS2026-07-06

Unified Audio Intelligence Without Regressing on Text Intelligence

Zhifeng Kong, Sang-gil Lee, Jaehyeon Kim, Boxin Wang, Zihan Liu, Sungwon Kim, et al.

Audio intelligence involves understanding, reasoning about, and generating both audio and speech. In this work, we introduce Nemotron-Labs-Audex-30B-A3B (Audex), a unified audio-text LLM built on Nemotron-Cascade-2-30B-A3B, a strong text-only MoE LLM. Audex adopts a simple unifie…

View free PDFSource page
arxiveess.AScs.AIcs.SD2026-07-15

Cover First, Disagree Softly: Rethinking Mismatch-First Active Learning for Frame-Level Audio Classification

Shiqi Zhang, Tuomas Virtanen

Sound event detection relies on frame-level strong labels whose annotation is expensive. Active learning addresses this problem by selecting the audio segments whose labels help the classifier most. One of the prevailing acquisition strategies for this task, mismatch-first farthe…

View free PDFSource page
arxivcs.SDcs.AI2026-07-05

Auto-AEG: Scalable Data Construction for Open-Vocabulary Audio Event Grounding

Zihan Zhang, Xize Cheng, Wenhao Yan, Tong Zhang, Dongjie Fu, Boyun Zhang, et al.

Large Audio-Language Models (LALMs) reason fluently about sound yet struggle to localize precisely when events occur, while classical Sound Event Detection attains frame-level precision only over a closed label set. At the intersection of these paradigms lies the task of Open-Voc…

View free PDFSource page