Concept · 8 episode(s)

Context Fatigue

← all concepts

Definition

Context fatigue is the observed degradation in model performance as conversation or document context grows, even within the nominal context window — needles get harder to find, instructions earlier in the prompt get less weight, reasoning gets sloppier. It’s why “long context” on the spec sheet doesn’t always mean “long context that works.”

Episodes covering this

  1. 198
    The Model That Knows the Answer and Can't Say It
    Can Language Models Actually Retrieve In-Context? Drowning in Documents at Million Token Scale
    Gollapudi, Gupta, Singhal et al. · UC Berkeley·17 min·Jul 03, 2026
  2. 164
    The Summarizer That Quietly Deletes Your Agent's Safety Rules
    Governance Decay: How Context Compaction Silently Erases Safety Constraints in Long-Horizon LLM Agents
    Chen · Beijing Institute of Technology·28 min·Jun 23, 2026
  3. 146
    How an Innocent README Can Freeze an AI Agent's Safety Check for an Hour
    From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails
    Zhou, Wang, Ma et al. · Hong Kong University of Science and Technology·26 min·Jun 15, 2026
  4. 142
    Training a Tiny Model to Run the Plumbing Between an Agent and the World
    HarnessBridge: Learnable Bidirectional Controller for LLM Agent Harness
    Wang, Wang, Taylor et al. · University of California·24 min·Jun 12, 2026
  5. 116
    Why Streaming Half a Reasoning Chain Beats Sending the Whole Thing
    Streaming Communication in Multi-Agent Reasoning
    Yang, Xu, Wang et al. · HKUST (GZ)·26 min·Jun 04, 2026
  6. 086
    Why Frozen-Weight Agents Still Get Worse Over Time
    Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems
    Zhu, Ro, Robertson et al. · The University of Texas at Austin·23 min·May 27, 2026
  7. 035
    Why Frontier Agents Ask for Clarification at Exactly the Wrong Moment
    Ask Early, Ask Late, Ask Right: When Does Clarification Timing Matter for Long-Horizon Agents?
    Gulati, Gupta, Lumer et al. · PricewaterhouseCoopers U.S.·29 min·May 11, 2026
  8. 017
    When the Agent Grades Its Own Homework: A Brutal New Benchmark for AI Workers
    Gym-Anything: Turn any Software into an Agent Environment
    Aggarwal, Neubig, Welleck · CMU·31 min·May 03, 2026

Worth reading next

Papers we haven't done a deep dive on yet, but would recommend on this topic.