Language Model Robustness

Follow this topic Ask about this topic

Momentum

35 papers in the last four weeks, up 46% on the four weeks before. 0.3% of all new papers.

Jul 13Week of Sep 28

Latest papers 338

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
Absurd World: A Simple Yet Powerful Method to Absurdify the Real-world for Probing LLM Reasoning Capabilities
May 10, 2026Ryan Albright, Golam Md Muktadir, Zarif Ikram +3LLM Evaluation Language Model Robustness
The Nueva School, San Mateo, USA · University of Southern California · Notre Dame College, Dhaka, Bangladesh +2
Can We Trust LLMs for Mental Health Screening? Consistency, ASR Robustness, and Evidence Faithfulness
May 10, 2026Erfan Loweimi, Sofia de la Fuente Garcia, Samira Loveymi +2Healthcare Mental Health
Usher Institute, Edinburgh Medical School, University of Edinburgh, Edinburgh, UK · Department of Computer Engineering, Ahvaz Campus, Islamic Azad University, Ahvaz, Iran · School of Health and Social Care, Edinburgh Napier University, Edinburgh, UK
How LLMs Are Persuaded: A Few Attention Heads, Rerouted
May 10, 2026Xiangkun Sun, Lingkai Kong, Aoqi Zhang +2Self-Attention LLM Interpretability
Northeastern University · Harvard University · Tsinghua University +1
WASIL: In-the-Wild Arabic Spoken Interactions with LLMs
May 9, 2026Zien Sheikh Ali, Hamdy Mubarak, Soon-Gyo Jung +3Voice Agent Evaluation Arabic NLP
Qatar Computing Research Institute, Qatar
FragileFlow: Spectral Control of Correct-but-Fragile Predictions for Foundation Model Robustness
May 9, 2026Zhuoyun Li, Boxuan Wang, Jinwei Hu +2VLM Robustness Spectral Regularization
School of Computer Science and Informatics, University of Liverpool, UK
DiagnosticIQ: A Benchmark for LLM-Based Industrial Maintenance Action Recommendation from Symbolic Rules
May 9, 2026Devin Yasith De Silva, Dhaval Patel, Christodoulos Constantinides +7Logical Reasoning LLM Evaluation
2Rensselaer Polytechnic Institute · 1IBM · 3Independent Contributor
Coordinates of Capability: A Unified MTMM-Geometric Framework for LLM Evaluation
May 8, 2026Adib Sakhawat, Tahsin Islam, Takia Farhin +3LLM Evaluation Language Model Robustness
Systems and Software Lab (SSL) Department of Computer Science and Engineering Islamic University of Technology, Dhaka, Bangladesh
Can Language Models Identify Side Effects of Breast Cancer Radiation Treatments?
May 8, 2026Natalie Seah, Danielle S. Bitterman, Daphna Spiegel +1Breast Cancer Healthcare
University of Virginia · Mass General Brigham, Dana-Farber Cancer Institute, Harvard Medical School
The Text Uncanny Valley: Non-Monotonic Performance Degradation in LLM Information Retrieval
May 8, 2026Zekai Tong, Ruiyao Xu, Aryan Shrivastava +2LLM Evaluation LLM Reliability
University of Chicago · 2Northwestern University
GSM-SEM: Benchmark and Framework for Generating Semantically Variant Augmentations
May 8, 2026Jyotika Singh, Fang Tu, Aziza Mirsaidova +11Mathematical Reasoning Benchmarks Benchmark Design
Oracle AI
Can LLMs Take Retrieved Information with a Grain of Salt?
May 7, 2026Behzad Shayegh, Mohamed Osama Ahmed, Fred Tung +1Retrieval-Augmented Generation LLM Evaluation
RBC Borealis
Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders
May 7, 2026Shunchang Liu, Xin Chen, Belen Martin Urcelay +1Reward Modeling Sparse Autoencoders
ETH Zürich · Georgia Institute of Technology · ELLIS Institute Finland +1
Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts
May 7, 2026Sneha Oram, Ojaswita Bhushan, Pushpak BhattacharyyaLLM Evaluation Language Model Robustness
Indian Institute of Technology Bombay, India
Information Theoretic Adversarial Training of Large Language Models
May 6, 2026Yiwei Zhang, Jeremiah Birrell, Reza Ebrahimi +3Adversarial Training LLM Alignment
Purdue University · West Lafayette, IN 47907, USA · Texas State University +5
Paraphrase-Induced Output-Mode Collapse: When LLMs Break Character Under Semantically Equivalent Inputs
May 6, 2026Aofan Liu, Jingxiang MengPrompt Sensitivity LLM Evaluation
1Peking University. · University of Chicago.
NoisyCausal: A Benchmark for Evaluating Causal Reasoning Under Structured Noise
May 5, 2026Zhi Xu, Yun FuStructural Causal Models Causal Reasoning in Language Models
Northeastern University
TriBench-Ko: Evaluating LLM Risks in Judicial Workflows
May 5, 2026Haesung Lee, Gyubin Choi, Eun-Ju Lee +5LLM Evaluation Language Model Safety Evaluation
Center for Trustworthy AI, Seoul National University · Graduate School of Data Science, Seoul National University · Responsible AI Team, KT +1
Stop Automating Peer Review Without Rigorous Evaluation
May 4, 2026Joachim Baumann, Jiaxin Pei, Sanmi Koyejo +1LLM-as-a-Judge Scholarly Peer Review
Stanford University · Bocconi University
Evaluating Reasoning Models for Queries with Presuppositions
May 4, 2026Rose Sathyanathan, Kinshuk Vasisht, Danish PruthiLLM Evaluation Language Model Robustness
Indian Institute of Science Bengaluru, KA, India
Perturbation Dose Responses in Recursive LLM Loops: Raw Switching, Stochastic Floors, and Persistent Escape under Append, Replace, and Dialog Updates
May 4, 2026Pawel KaplanskiLanguage Model Robustness LLM Context Management
Kaplanski AI Lab · Kaplanski Ai Lab
What Single-Prompt Accuracy Misses: A Multi-Variant Reliability Audit of Language Models
May 3, 2026Ranit Karmakar, Jayita ChatterjeeLLM Evaluation Language Model Calibration
Harvard University, Boston, MA 02115 · Boston, MA 02134
Format Sensitivity Index: Token-Controlled Prompt Wrapper Robustness and Schema Compliance in LLM Benchmarking
May 2, 2026Deep Pankajbhai MehtaPrompt Sensitivity LLM Evaluation
Adobe Inc.
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
Lost in the Tower of Babel: The Adverse Effects of Incidental Multilingualism in LLMs
May 2, 2026Anjishnu Mukherjee, Chutong Meng, Antonios AnastasopoulosMultilingual Language Models Multilingual Language Model Evaluation
Department of Computer Science George Mason University
When Embedding-Based Defenses Fail: Rethinking Safety in LLM-Based Multi-Agent Systems
May 1, 2026Lingxi Zhang, Guangtao Zheng, Hanjie ChenAdversarial Attacks Language Model Robustness
1Rice University · University of Virginia
CLEAR: Revealing How Noise and Ambiguity Degrade Reliability in LLMs for Medicine
May 1, 2026Kevin H. Guo, Chao Yan, Avinash Baidya +5Healthcare LLM Reliability
Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA · Intuit AI Research, Mountain View, CA, USA · Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, USA +2
Are Tools All We Need? Unveiling the Tool-Use Tax in LLM Agents
Apr 30, 2026Kaituo Zhang, Zhen Xiong, Mingyu Zhong +4Tool-Augmented Language Model Agents Agentic Reasoning
University of Houston · University of Southern California · 3New York University +2
HealthBench Professional: Evaluating Large Language Models on Real Clinician Chats
Apr 30, 2026Rebecca Soskin Hicks, Mikhail Trofimov, Dominick Lim +13Healthcare Clinical Decision Support
OpenAI
A Reproducibility Analysis of PO4ISR: Diagnosing and Mitigating Semantic Drift in LLM-Based Session Recommendation
Apr 29, 2026Aditya Tiwari, Konduri Naga Lakshmi Rekha, Rajesh Kumar MundotiyaSequential Recommendation LLM Prompting
MATRA Lab Department of Computer Science and Engineering Indian Institute of Technology Bhilai Chhattisgarh, India 491002

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 338