Cover Image

Why Retrying Fails: Context Contamination in LLM Agent Pipelines

Fuente: arXiv

Saved in:

Bibliographic Details
Main Author:	Yang, Zhanfu
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Why Do Multi-Agent LLM Systems Fail?
by: Cemri, Mert, et al.
Published: (2025)

Retrying vs Resampling in AI Control
by: Lucassen, James, et al.
Published: (2026)

Why Do LLM-based Web Agents Fail? A Hierarchical Planning Perspective
by: Aghzal, Mohamed, et al.
Published: (2026)

Why Do Transformers Fail to Forecast Time Series In-Context?
by: Zhou, Yufa, et al.
Published: (2025)

WebSuite: Systematically Evaluating Why Web Agents Fail
by: Li, Eric, et al.
Published: (2024)

When Refusals Fail: Unstable Safety Mechanisms in Long-Context LLM Agents
by: Hadeliya, Tsimur, et al.
Published: (2025)

Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness
by: Denisov-Blanch, Yegor, et al.
Published: (2026)

Why LLMs Fail at Causal Discovery and How Interventional Agents Escape
by: Roy, Amartya, et al.
Published: (2026)

Where LLM Agents Fail and How They can Learn From Failures
by: Zhu, Kunlun, et al.
Published: (2025)

Emergence of Exploration in Policy Gradient Reinforcement Learning via Retrying
by: Nishimori, Soichiro, et al.
Published: (2026)

State Contamination in Memory-Augmented LLM Agents
by: Wang, Yian, et al.
Published: (2026)

Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
by: Wang, Zehong, et al.
Published: (2026)

Why Chain of Thought Fails in Clinical Text Understanding
by: Wu, Jiageng, et al.
Published: (2025)

Coherence Collapse: Diagnosing Why Code Agents Fail After Reaching the Right Code
by: Kim, Myeongsoo, et al.
Published: (2026)

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

The Keyhole Effect: Why Chat Interfaces Fail at Data Analysis
by: Reddy, Mohan
Published: (2026)

Why Do AI Agents Systematically Fail at Cloud Root Cause Analysis?
by: Kim, Taeyoon, et al.
Published: (2026)

SkillGenBench: Benchmarking Skill Generation Pipelines for LLM Agents
by: Zhou, Yifan, et al.
Published: (2026)

LLM-Human Pipeline for Cultural Context Grounding of Conversations
by: Pujari, Rajkumar, et al.
Published: (2024)

When Agents Fail to Act: A Diagnostic Framework for Tool Invocation Reliability in Multi-Agent LLM Systems
by: Huang, Donghao, et al.
Published: (2026)

Escaping the Context Bottleneck: Active Context Curation for LLM Agents via Reinforcement Learning
by: Li, Xiaozhe, et al.
Published: (2026)

Why Retrieval-Augmented Generation Fails: A Graph Perspective
by: Guo, Kai, et al.
Published: (2026)

The Defense Trilemma: Why Prompt Injection Defense Wrappers Fail?
by: Bhatt, Manish, et al.
Published: (2026)

AutoContext: Instance-Level Context Learning for LLM Agents
by: Cai, Kuntai, et al.
Published: (2025)

No Attacker Needed: Unintentional Cross-User Contamination in Shared-State LLM Agents
by: Yang, Tiankai, et al.
Published: (2026)

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention
by: Qiu, Haiquan, et al.
Published: (2025)

Traceability and Accountability in Role-Specialized Multi-Agent LLM Pipelines
by: Barrak, Amine
Published: (2025)

Galton's Law of Mediocrity: Why Large Language Models Regress to the Mean and Fail at Creativity in Advertising
by: Keon, Matt, et al.
Published: (2025)

PRISM: Generation-Time Detection and Mitigation of Secret Leakage in Multi-Agent LLM Pipelines
by: Tapwal, Riya, et al.
Published: (2026)

Standard Benchmarks Fail -- Auditing LLM Agents in Finance Must Prioritize Risk
by: Chen, Zichen, et al.
Published: (2025)

Better Eyes, Better Thoughts: Why Vision Chain-of-Thought Fails in Medicine
by: Wu, Yuan, et al.
Published: (2026)

R$^3$L: Reflect-then-Retry Reinforcement Learning with Language-Guided Exploration, Pivotal Credit, and Positive Amplification
by: Shi, Weijie, et al.
Published: (2026)

Five Fatal Assumptions: Why T-Shirt Sizing Systematically Fails for AI Projects
by: Soundaramourty, Raja, et al.
Published: (2026)

Why Attention Fails: A Taxonomy of Faults in Attention-Based Neural Networks
by: Jahan, Sigma, et al.
Published: (2025)

LM Agents May Fail to Act on Their Own Risk Knowledge
by: Tang, Yuzhi, et al.
Published: (2025)

Why Federated Optimization Fails to Achieve Perfect Fitting? A Theoretical Perspective on Client-Side Optima
by: Lei, Zhongxiang, et al.
Published: (2025)

LLM Self-Explanations Fail Semantic Invariance
by: Szeider, Stefan
Published: (2026)

RetriBooru: Leakage-Free Retrieval of Conditions from Reference Images for Subject-Driven Generation
by: Tang, Haoran, et al.
Published: (2023)

Agent Benchmarks Fail Public Sector Requirements
by: Rystrøm, Jonathan, et al.
Published: (2026)

SlimPipe: Memory-Thrifty and Efficient Pipeline Parallelism for Long-Context LLM Training
by: Li, Zhouyang, et al.
Published: (2025)