Lost in Compaction: Evaluating Side-Constraint Loss under Context Compaction
Source
Zhiqi Wang, Yichi Zhang, Dongwon Lee, Yuchen Yang
Author
Zhiqi Wang, Yichi Zhang, Dongwon Lee, Yuchen Yang
Published
Terms in this piece · Glossary
context compaction — Summarizing an agent's earlier conversation to free room in the context window so a long session can keep going.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
Why it matters
Any long-running AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → that compacts context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → can quietly discard the safety instruction a user gave it earlier in the session, and this quantifies how often that happens rather than leaving it as folklore.
articleBetter, Faster, Stronger: Programmatic Skill Learning Best Reduces Agent CostZixi Huang, Xiheng Wang, Andrew Wang, William Jurayj, Bernal Jim\'enez Guti\'errez, Daniel Khashabi, Nicholas Andrews
articleHarnessing LLMs for Document-Guided Fuzzing of Python LibrariesBin Duan, Tarek Mahmud, Meiru Che, Yan Yan, Naipeng Dong, Dan Dongseong Kim, Guowei YangarticleEvaluating Context Compression for AI AgentsFactory