Cover Image

When Scaling Fails: Mitigating Audio Perception Decay of LALMs via Multi-Step Perception-Aware Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Mao, Ruixiang, Ma, Xiangnan, Chen, Dan, Zhu, Ziming, Ge, Yuan, Hao, Aokai, Zhao, Haishu, Huo, Yifu, Yang, Qing, Chang, Kaiyan, Liu, Xiaoqian, Wang, Chenglong, He, Qiaozhi, Xiao, Tong, Zhu, Jingbo
Format:	Preprint
Published:	2026
Subjects:	Sound Artificial Intelligence Audio and Speech Processing
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

APR: Penalizing Structural Redundancy in Large Reasoning Models via Anchor-based Process Rewards
by: Chang, Kaiyan, et al.
Published: (2026)

StyleBench: Evaluating Speech Language Models on Conversational Speaking Style Control
by: Zhao, Haishu, et al.
Published: (2026)

On the Emotion Understanding of Synthesized Speech
by: Ge, Yuan, et al.
Published: (2026)

M-CIF: Multi-Scale Alignment For CIF-Based Non-Autoregressive ASR
by: Mao, Ruixiang, et al.
Published: (2025)

Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models
by: Chang, Kaiyan, et al.
Published: (2025)

SPS: Steering Probability Squeezing for Better Exploration in Reinforcement Learning for Large Language Models
by: Huo, Yifu, et al.
Published: (2026)

Efficient Prompting Methods for Large Language Models: A Survey
by: Chang, Kaiyan, et al.
Published: (2024)

LRHP: Learning Representations for Human Preferences via Preference Pairs
by: Wang, Chenglong, et al.
Published: (2024)

SERM: Self-Evolving Relevance Model with Agent-Driven Learning from Massive Query Streams
by: Wang, Chenglong, et al.
Published: (2026)

MSRL: Scaling Generative Multimodal Reward Modeling via Multi-Stage Reinforcement Learning
by: Wang, Chenglong, et al.
Published: (2026)

GRAM: A Generative Foundation Reward Model for Reward Generalization
by: Wang, Chenglong, et al.
Published: (2025)

EvA: An Evidence-First Audio Understanding Paradigm for LALMs
by: Xie, Xinyuan, et al.
Published: (2026)

RankPrompt: Step-by-Step Comparisons Make Language Models Better Reasoners
by: Hu, Chi, et al.
Published: (2024)

LaTeXTrans: Structured LaTeX Translation with Multi-Agent Coordination
by: Zhu, Ziming, et al.
Published: (2025)

RoVRM: A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data
by: Wang, Chenglong, et al.
Published: (2024)

CAF-Score: Calibrating CLAP with LALMs for Reference-free Audio Captioning Evaluation
by: Lee, Insung, et al.
Published: (2026)

DaPT: A Dual-Path Framework for Multilingual Multi-hop Question Answering
by: Wang, Yilin, et al.
Published: (2026)

Probing Preference Representations: A Multi-Dimensional Evaluation and Analysis Method for Reward Models
by: Wang, Chenglong, et al.
Published: (2025)

Can Large Audio Language Models Understand Audio Well? Speech, Scene and Events Understanding Benchmark for LALMs
by: Yin, Han, et al.
Published: (2025)

GRAM-R$^2$: Self-Training Generative Foundation Reward Models for Reward Reasoning
by: Wang, Chenglong, et al.
Published: (2025)

When Reward Hacking Rebounds: Understanding and Mitigating It with Representation-Level Signals
by: Wu, Rui, et al.
Published: (2026)

Benchmarking and Confidence Evaluation of LALMs For Temporal Reasoning
by: Bhattacharya, Debarpan, et al.
Published: (2025)

A Modular-based Strategy for Mitigating Gradient Conflicts in Simultaneous Speech Translation
by: Liu, Xiaoqian, et al.
Published: (2024)

A Deep Learning Approach for Augmenting Perceptional Understanding of Histopathology Images
by: Hu, Xiaoqian
Published: (2025)

Sobolev Differentiability Properties of Logarithmic Modulus of Real Analytic Functions
by: Shi, Ziming, et al.
Published: (2022)

Mind the Hitch: Dynamic Calibration and Articulated Perception for Autonomous Trucks
by: Zhu, Morui, et al.
Published: (2026)

Hybrid Alignment Training for Large Language Models
by: Wang, Chenglong, et al.
Published: (2024)

HEAL: A Hypothesis-Based Preference-Aware Analysis Framework
by: Huo, Yifu, et al.
Published: (2025)

Human Perception of Audio Deepfakes
by: Müller, Nicolas M., et al.
Published: (2021)

Boosting Text-To-Image Generation via Multilingual Prompting in Large Multimodal Models
by: Mu, Yongyu, et al.
Published: (2025)

UMI-3D: Extending Universal Manipulation Interface from Vision-Limited to 3D Spatial Perception
by: Wang, Ziming
Published: (2026)

When Large Language Model Agents Meet 6G Networks: Perception, Grounding, and Alignment
by: Xu, Minrui, et al.
Published: (2024)

Exploring Autonomous Agents: A Closer Look at Why They Fail When Completing Tasks
by: Lu, Ruofan, et al.
Published: (2025)

PRJ: Perception-Retrieval-Judgement for Generated Images
by: Fu, Qiang, et al.
Published: (2025)

VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset
by: Liu, Jing, et al.
Published: (2023)

Audio-VLA: Adding Contact Audio Perception to Vision-Language-Action Model for Robotic Manipulation
by: Wei, Xiangyi, et al.
Published: (2025)

RouteLMT: Learned Sample Routing for Hybrid LLM Translation Deployment
by: Luo, Yingfeng, et al.
Published: (2026)

Learning Evaluation Models from Large Language Models for Sequence Generation
by: Wang, Chenglong, et al.
Published: (2023)

Slow Perception: Let's Perceive Geometric Figures Step-by-step
by: Wei, Haoran, et al.
Published: (2024)

Audio-Guided Visual Perception for Audio-Visual Navigation
by: Wang, Yi, et al.
Published: (2025)