Noise Robustness

Follow this topic Ask about this topic

Momentum

3 papers in the last four weeks, down 40% on the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 71

All topics

EchoDistill: Robust Large Audio Language Models via Noisy-to-Clean Self-Distillation
May 11, 2026Kaiwen Luo, Chunxi Luo, Liang Lin +5Audio Understanding Audio-Language Models
NTU · SHU · ICT, CAS +4
Where Do Prompt Perturbations Break Generation? A Segment-Level View of Robustness in LoRA-Tuned Language Models
May 2, 2026Zhuoyun Li, Boxuan Wang, Jinwei Hu +6Neural Network Robustness Noise Robustness
School of Computer Science and Informatics, University of Liverpool, UK
Laplace-Bridged Randomized Smoothing for Fast Certified Robustness
Apr 27, 2026Miao Lin, MD Saifur Rahman Mazumder, Feng Yu +2Randomized Smoothing Neural Network Robustness
Department of Computer Science, Old Dominion University, Norfolk, VA, USA · Department of Mathematical Sciences, University of Texas at El Paso, El Paso, TX, USA · School of Cybersecurity, Old Dominion University, Norfolk, VA, USA
CCAR: Intrinsic Robustness as an Emergent Geometric Property
Apr 18, 2026Akash Samanta, Manish Pratap Singh, Debasis ChaudhuriDisentangled Representation Learning Representation Learning
Techno India University, Salt Lake, Kolkata, India · DRDO Young Scientist Laboratory - CT, Chennai, India
AstroSURE: Learning to Remove Noise from Astronomical Images Without Ground Truth Data
Apr 18, 2026Omid Vaheb, Sebastien Fabbro, Stark DraperSelf-Supervised Image Denoising Noise Robustness
Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, 10 King’s College Road, Toronto, ON M5S 3G4, Canada · National Research Council Herzberg Astronomy an Astrophysics Research Centre, 5071 West Saanich Road, Victoria, BC V9E 2E7, Canada · Department of Computer Science, University of British Columbia, Vancouver, BC, V6T 1Z4, Canada
Learning to Recorrupt: Noise Distribution Agnostic Self-Supervised Image Denoising
Mar 26, 2026Brayan Monroy, Jorge Bacca, Julián TachellaSelf-Supervised Image Denoising Self-Supervised Learning
Universidad Industrial de Santander, Colombia · CNRS, ENS de Lyon, France
Focus Then Listen: An Empirical Study of Plug-and-Play Audio Enhancer for Noise-Robust Large Audio Language Models
Mar 5, 2026Han Yin, Yang Xiao, Younghoo Kwon +2Audio Understanding Audio-Language Models
School of Electrical Engineering, KAIST, Daejeon, Republic of Korea · University of Melbourne, Australia
SQuTR: A Robustness Benchmark for Spoken Query to Text Retrieval under Acoustic Noise
Feb 13, 2026Yuejie Li, Ke Yang, Yueying Hua +4Noise Robustness IR Evaluation
Ant Group Hangzhou, China · Soochow University Suzhou, China · University of Science and Technology of China Hefei, China +3
Dual Randomized Smoothing: Beyond Global Noise Variance
Dec 1, 2025Chenhao Sun, Yuhao Mao, Martin VechevRandomized Smoothing Neural Network Robustness
Department of Computer Science, ETH Zürich, Switzerland
DHAuDS: A Dynamic and Heterogeneous Audio Benchmark for Test-Time Adaptation
Nov 23, 2025Weichuang Shao, Iman Yi Liao, Tomas Henrique Bode Maul +1Test-Time Adaptation Neural Network Robustness
School of Computer and Mathematical Sciences, University of Nottingham Malaysia, Semenyih, Malaysia
Topological data analysis using persistent discrete homology
Jun 17, 2025Chris Kapulkin, Nathan KershawNoise Robustness Persistent Homology

May 11, 2026, cs.CLJ/K move · Enter open · S save

EchoDistill: Robust Large Audio Language Models via Noisy-to-Clean Self-Distillation

Kaiwen Luo, Chunxi Luo, Liang Lin, Yuxuan Li+4

NTU · SHU · ICT, CAS+4

Audio Understanding Audio-Language Models Self-Distillation Noise Robustness Language Model Robustness

Open paper AskPDF arXiv

Large Audio Language Models (LALMs) remain vulnerable to acoustic noise, which can obscure task-relevant evidence and produce unreliable responses. We propose EchoDistill, a noisy-to-clean self-distillation framework that uses clean audio as privileged information during post-training. A noisy-input student samples candidate responses reflecting its inference-time behavior, while a frozen copy of the same backbone processes the corresponding clean audio. EchoDistill combines masked response-token distillation, task-gated consistency shaping, and teacher-referenced group-relative optimization to align noisy-input generation with clean-conditioned semantics. Only the student is retained at inference time, introducing no additional inference cost. Across three LALM backbones and three audio domains at -10dB, EchoDistill improves average noisy-input accuracy by 1.63 percentage points over the strongest baseline. On Qwen2.5-Omni, it raises noisy-input accuracy from 59.33% to 62.94%, while clean-audio accuracy increases from 76.56% to 77.56%. Replacing matched audio with random, shuffled, or silent inputs reduces accuracy by 3.08-6.42 points, confirming that matched acoustic evidence contributes to its predictions. Additional evaluations show improvements on held-out additive noises and external benchmarks, while revealing that these gains do not reliably extend to non-additive distortions. These results demonstrate robust post-training improvements under severe additive noise without sacrificing clean-audio capability across diverse tasks.

Momentum

Related topics

Latest papers 71