Image de couverture de livre

When Do LLM Agents Treat Surface Noise Differently from Semantic Noise? A 68-Cell Measurement Study with a Held-Out Trace-Level Validation

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhang, Liyun, Guo, Jiayi
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

When Does Divide and Conquer Work for Long Context LLM? A Noise Decomposition Framework
par: Xu, Zhen, et autres
Publié: (2025)

STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?
par: Chao, Hanxiang, et autres
Publié: (2026)

Do Lexical and Contextual Coreference Resolution Systems Degrade Differently under Mention Noise? An Empirical Study on Scientific Software Mentions
par: Alkan, Atilla Kaan, et autres
Publié: (2026)

When Safety Blocks Sense: Measuring Semantic Confusion in LLM Refusals
par: Anonto, Riad Ahmed, et autres
Publié: (2025)

Incorporating Error Level Noise Embedding for Improving LLM-Assisted Robustness in Persian Speech Recognition
par: Rahmani, Zahra, et autres
Publié: (2025)

Weed Out, Then Harvest: Dual Low-Rank Adaptation is an Effective Noisy Label Detector for Noise-Robust Learning
par: Yuan, Bo, et autres
Publié: (2025)

From Noise to Signal: When Outliers Seed New Topics
par: Zve, Evangelia, et autres
Publié: (2026)

Measuring the Effect of Transcription Noise on Downstream Language Understanding Tasks
par: Shapira, Ori, et autres
Publié: (2025)

When TableQA Meets Noise: A Dual Denoising Framework for Complex Questions and Large-scale Tables
par: Ye, Shenghao, et autres
Publié: (2025)

Do Proactive Agents Really Need an LLM to Decide When to Wake and What to Anchor?
par: Liu, Xiaoze, et autres
Publié: (2026)

LLM NL2SQL Robustness: Surface Noise vs. Linguistic Variation in Traditional and Agentic Settings
par: Tu, Lifu, et autres
Publié: (2026)

Thinking Out Loud: Do Reasoning Models Know When They're Right?
par: Zeng, Qingcheng, et autres
Publié: (2025)

PersonaTrace: Synthesizing Realistic Digital Footprints with LLM Agents
par: Wang, Minjia, et autres
Publié: (2026)

Noise or Nuance: An Investigation Into Useful Information and Filtering For LLM Driven AKBC
par: Clay, Alex, et autres
Publié: (2025)

Cutting Through the Noise: Boosting LLM Performance on Math Word Problems
par: Anantheswaran, Ujjwala, et autres
Publié: (2024)

StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs
par: Song, Yuhan, et autres
Publié: (2025)

Noise-BERT: A Unified Perturbation-Robust Framework with Noise Alignment Pre-training for Noisy Slot Filling Task
par: Zhao, Jinxu, et autres
Publié: (2024)

When Relations Break: Analyzing Relation Hallucination in Vision-Language Model Under Rotation and Noise
par: Shin, Philip Wootaek, et autres
Publié: (2026)

ToolFlood: Beyond Selection -- Hiding Valid Tools from LLM Agents via Semantic Covering
par: Jawad, Hussein, et autres
Publié: (2026)

Reinforcement Learning for LLM-based Multi-Agent Systems through Orchestration Traces
par: Zhang, Chenchen
Publié: (2026)

Text-as-Signal: Quantitative Semantic Scoring with Embeddings, Logprobs, and Noise Reduction
par: Moreira, Hugo
Publié: (2026)

Don't Blame the Data, Blame the Model: Understanding Noise and Bias When Learning from Subjective Annotations
par: Anand, Abhishek, et autres
Publié: (2024)

Beyond the Surface: Measuring Self-Preference in LLM Judgments
par: Chen, Zhi-Yuan, et autres
Publié: (2025)

Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?
par: Li, Xinzhe, et autres
Publié: (2023)

SafeSearch: Do Not Trade Safety for Utility in LLM Search Agents
par: Zhan, Qiusi, et autres
Publié: (2025)

When Agents Trade: Live Multi-Market Trading Benchmark for LLM Agents
par: Qian, Lingfei, et autres
Publié: (2025)

Provable Defense Framework for LLM Jailbreaks via Noise-Augumented Alignment
par: Cheng, Zehua, et autres
Publié: (2026)

Explainable Token-level Noise Filtering for LLM Fine-tuning Datasets
par: Yang, Yuchen, et autres
Publié: (2026)

Do Cognitively Interpretable Reasoning Traces Improve LLM Performance?
par: Bhambri, Siddhant, et autres
Publié: (2025)

DoCIA: An Online Document-Level Context Incorporation Agent for Speech Translation
par: Lyu, Xinglin, et autres
Publié: (2025)

When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning
par: Wei, Jiaqi, et autres
Publié: (2026)

Noise-powered Multi-modal Knowledge Graph Representation Framework
par: Chen, Zhuo, et autres
Publié: (2024)

Semantic Role Labeling Guided Out-of-distribution Detection
par: Zou, Jinan, et autres
Publié: (2023)

RNG: Reducing Multi-level Noise and Multi-grained Semantic Gap for Joint Multimodal Aspect-Sentiment Analysis
par: Liu, Yaxin, et autres
Publié: (2024)

Noise-Robust De-Duplication at Scale
par: Silcock, Emily, et autres
Publié: (2022)

Noise Steering for Controlled Text Generation: Improving Diversity and Reading-Level Fidelity in Arabic Educational Story Generation
par: Khalid, Haziq Mohammad, et autres
Publié: (2026)

How Much Heavy Lifting Can an Agent Harness Do?: Measuring the LLM's Residual Role in a Planning Agent
par: Jung, Sungwoo, et autres
Publié: (2026)

What Do VLMs NOTICE? A Mechanistic Interpretability Pipeline for Gaussian-Noise-free Text-Image Corruption and Evaluation
par: Golovanevsky, Michal, et autres
Publié: (2024)

HEART-Bench: Do LLM Agents Exhibit Human-like Psychology?
par: Peng, Weihan, et autres
Publié: (2026)

RAVEL: Reasoning Agents for Validating and Evaluating LLM Text Synthesis
par: Feng, Andrew Zhuoer, et autres
Publié: (2026)