Escape the Language Prior: Mitigating Late-Stage Modality Collapse in Audio Reasoning via Modality-Aware Policy Optimization

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Xiao, Cihan, Shao, Yiwen, Li, Chenxing, He, Xiang, Liang, Zhenwen, Yves, Steve, Khudanpur, Sanjeev, Bo, Liefeng
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

SpatialEmb: Extract and Encode Spatial Information for 1-Stage Multi-channel Multi-speaker ASR on Arbitrary Microphone Arrays
di: Shao, Yiwen, et al.
Pubblicazione: (2026)

Unsupervised Speech Enhancement using Data-defined Priors
di: Klement, Dominik, et al.
Pubblicazione: (2025)

Generalizing from SIMPLE to HARD Visual Reasoning: Can We Mitigate Modality Imbalance in VLMs?
di: Park, Simon, et al.
Pubblicazione: (2025)

HENT-SRT: Hierarchical Efficient Neural Transducer with Self-Distillation for Joint Speech Recognition and Translation
di: Hussein, Amir, et al.
Pubblicazione: (2025)

Segment-Aligned Policy Optimization for Multi-Modal Reasoning
di: Gao, Lei, et al.
Pubblicazione: (2026)

Separate First, Fuse Later: Mitigating Cross-Modal Interference in Audio-Visual LLMs Reasoning with Modality-Specific Chain-of-Thought
di: Li, Xuanchen, et al.
Pubblicazione: (2026)

Adversarial Attacks and Defenses for Speech Recognition Systems
di: Żelasko, Piotr, et al.
Pubblicazione: (2021)

Mitigating Modal Imbalance in Multimodal Reasoning
di: Wu, Chen Henry, et al.
Pubblicazione: (2025)

Reliable Multi-Modal Object Re-Identification via Modality-Aware Graph Reasoning
di: Wan, Xixi, et al.
Pubblicazione: (2025)

Multi-Channel Multi-Speaker ASR Using Target Speaker's Solo Segment
di: Shao, Yiwen, et al.
Pubblicazione: (2024)

TSAK: Two-Stage Semantic-Aware Knowledge Distillation for Efficient Wearable Modality and Model Optimization in Manufacturing Lines
di: Bello, Hymalai, et al.
Pubblicazione: (2024)

Dimensionality-Aware Anomaly Detection in Learned Representations of Self-Supervised Speech Models
di: Arcos-Holzinger, Sandra, et al.
Pubblicazione: (2026)

Audio-DeepThinker: Progressive Reasoning-Aware Reinforcement Learning for High-Quality Chain-of-Thought Emergence in Audio Language Models
di: He, Xiang, et al.
Pubblicazione: (2026)

Beyond Semantic Priors: Mitigating Optimization Collapse for Generalizable Visual Forensics
di: Liu, Jipeng, et al.
Pubblicazione: (2026)

AudioGenie-Reasoner: A Training-Free Multi-Agent Framework for Coarse-to-Fine Audio Deep Reasoning
di: Rong, Yan, et al.
Pubblicazione: (2025)

Modality Equilibrium Matters: Minor-Modality-Aware Adaptive Alternating for Cross-Modal Memory Enhancement
di: Shi, Xiang, et al.
Pubblicazione: (2025)

Collapsing Constructive and Intuitionistic Modal Logics
di: Pacheco, Leonardo
Pubblicazione: (2024)

Audio-Guided Dynamic Modality Fusion with Stereo-Aware Attention for Audio-Visual Navigation
di: Li, Jia, et al.
Pubblicazione: (2025)

Pay More Attention To Audio: Mitigating Imbalance of Cross-Modal Attention in Large Audio Language Models
di: Wang, Junyu, et al.
Pubblicazione: (2025)

Advantage Collapse in Group Relative Policy Optimization: Diagnosis and Mitigation
di: He, Xixiang, et al.
Pubblicazione: (2026)

Connect, Collapse, Corrupt: Learning Cross-Modal Tasks with Uni-Modal Data
di: Zhang, Yuhui, et al.
Pubblicazione: (2024)

Multi-Stage Verification-Centric Framework for Mitigating Hallucination in Multi-Modal RAG
di: Chen, Baiyu, et al.
Pubblicazione: (2025)

Improving Neural Biasing for Contextual Speech Recognition by Early Context Injection and Text Perturbation
di: Huang, Ruizhe, et al.
Pubblicazione: (2024)

AVRT: Audio-Visual Reasoning Transfer through Single-Modality Teachers
di: Araujo, Edson, et al.
Pubblicazione: (2026)

OmniVideo-R1: Reinforcing Audio-visual Reasoning with Query Intention and Modality Attention
di: Chen, Zhangquan, et al.
Pubblicazione: (2026)

Infeasibility Aware Large Language Models for Combinatorial Optimization
di: Wang, Yakun, et al.
Pubblicazione: (2026)

CoGenAV: Versatile Audio-Visual Representation Learning via Contrastive-Generative Synchronization
di: Bai, Detao, et al.
Pubblicazione: (2025)

Residual Structure Theory in Modal Collapse Systems
di: Li, Y.Y.N
Pubblicazione: (2025)

Taming Modality Entanglement in Continual Audio-Visual Segmentation
di: Hong, Yuyang, et al.
Pubblicazione: (2025)

SimpleDoc: Multi-Modal Document Understanding with Dual-Cue Page Retrieval and Iterative Refinement
di: Jain, Chelsi, et al.
Pubblicazione: (2025)

Enhancing Multi-Modal LLMs Reasoning via Difficulty-Aware Group Normalization
di: Li, Jinghan, et al.
Pubblicazione: (2026)

SaSR-Net: Source-Aware Semantic Representation Network for Enhancing Audio-Visual Question Answering
di: Yang, Tianyu, et al.
Pubblicazione: (2024)

RMD: Robust Modal Decomposition with Constrained Bandwidth
di: Hao, Wang, et al.
Pubblicazione: (2025)

MiMIC: Mitigating Visual Modality Collapse in Universal Multimodal Retrieval While Avoiding Semantic Misalignment
di: Li, Juan, et al.
Pubblicazione: (2026)

StutterFuse: Mitigating Modality Collapse in Stuttering Detection with Jaccard-Weighted Metric Learning and Gated Fusion
di: Singh, Guransh, et al.
Pubblicazione: (2025)

DeepAudio-V1:Towards Multi-Modal Multi-Stage End-to-End Video to Speech and Audio Generation
di: Zhang, Haomin, et al.
Pubblicazione: (2025)

CRMD: Complex Robust Modal Decomposition
di: Hao, Wang, et al.
Pubblicazione: (2025)

Group Distributionally Robust Optimization-Driven Reinforcement Learning for LLM Reasoning
di: Panaganti, Kishan, et al.
Pubblicazione: (2026)

Modality-Aware Infrared and Visible Image Fusion with Target-Aware Supervision
di: Sun, Tianyao, et al.
Pubblicazione: (2025)

A Systematic Study of Cross-Modal Typographic Attacks on Audio-Visual Reasoning
di: Chen, Tianle, et al.
Pubblicazione: (2026)