# When Can Agents Forget Their Reasoning? ICLR for Long-Horizon Agent Context Compression

arXiv: https://arxiv.org/abs/2609.29875 · HTML: https://mlsanity.com/paper/2609.29875/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 24 of its 50 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)
- LLM agents and tool use / [Agent memory systems](https://mlsanity.com/agents/agent-memory-systems/)
- Efficient LLM inference and serving / [Efficient reasoning and agent inference](https://mlsanity.com/inference/efficient-reasoning-and-agent-inference/)

## Builds on (most cited first, 24 of 24)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [Discovering Latent Knowledge in Language Models Without Supervision](https://mlsanity.com/paper/2212.03827/) (2022)
- [Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking](https://mlsanity.com/paper/2403.09629/) (2024)
- [Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention](https://mlsanity.com/paper/2404.07143/) (2024)
- [Scaling Long-Horizon LLM Agent via Context-Folding](https://mlsanity.com/paper/2510.11967/) (2025)
- [ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization](https://mlsanity.com/paper/2509.13313/) (2025)
- [ACON: Optimizing Context Compression for Long-horizon LLM Agents](https://mlsanity.com/paper/2510.00615/) (2025)
- [AgentFold: Long-Horizon Web Agents with Proactive Context Management](https://mlsanity.com/paper/2510.24699/) (2025)
- [SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents](https://mlsanity.com/paper/2601.16746/) (2026)
- [Beyond Static Summarization: Proactive Memory Extraction for LLM Agents](https://mlsanity.com/paper/2601.04463/) (2026)
- [Active Context Compression: Autonomous Memory Management in LLM Agents](https://mlsanity.com/paper/2601.07190/) (2026)
- [ContextBudget: Budget-Aware Context Management for Long-Horizon Search Agents](https://mlsanity.com/paper/2604.01664/) (2026)
- [Self-Compacting Language Model Agents](https://mlsanity.com/paper/2606.23525/) (2026)
- [Self-GC: Self-Governing Context for Long-Horizon LLM Agents](https://mlsanity.com/paper/2607.00692/) (2026)
- [SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent](https://mlsanity.com/paper/2605.24468/) (2026)
- [ACM: Agentic Context Management for Long Horizon Tasks](https://mlsanity.com/paper/2607.23809/) (2026)
- [Beyond Compaction: Structured Context Eviction for Long-Horizon Agents](https://mlsanity.com/paper/2606.11213/) (2026)
- [Chain Of Thought Compression: A Theoretical Analysis](https://mlsanity.com/paper/2601.21576/) (2026)
- [CoMem: Context Management with A Decoupled Long-Context Model](https://mlsanity.com/paper/2605.30842/) (2026)
- [Context as an Environment: Programmatic Context Management for Long-Horizon Agents](https://mlsanity.com/paper/2608.21690/) (2026)

## Abstract (arXiv)

Long horizon language model agents continually accumulate reasoning history, increasing context length and inference cost even after earlier decisions have been executed and observed. Unlike static Chain of Thought compression, removing historical reasoning can change future actions and the resulting interaction trajectory. We study when such reasoning can be safely forgotten. We propose Interaction Aware Compression for Long Horizon Reasoning (ICLR), a training free online method that ranks reasoning blocks using frozen proxy entropy while preserving actions, tool calls, and observations. On 260 WorkBuddyBench tasks, ICLR improves average reward from 0.699 to 0.718, while reducing input, output, and cache read tokens by 25.5%, 14.4%, and 33.3%, respectively. Ablations reveal trajectory amplification, where local reasoning deletion produces nonlinear changes in total computation by altering subsequent interaction. Representation probing, activation patching, and controlled trajectory analyses further suggest that historical reasoning becomes more replaceable once task relevant derived state has been reliably externalized into code, files, tool outputs, or environmental feedback. These results characterize agent reasoning as dynamic working state rather than permanent interaction history.