SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent

• URL: http://arxiv.org/abs/2605.24468v1
• Date: Sat, 23 May 2026 08:37:16 GMT
• Title: SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent
• Abstract summary: Long-horizon agentic reasoning requires large language models to act over long interaction histories containing thoughts, tool calls, observations, and partial conclusions.<n>Existing approaches address this difficulty by truncating the interaction history, compressing it into shorter surrogates, or retrieving selected parts of it for reuse.<n>We propose State-Adaptive Memory, a standalone framework that consolidates ongoing interaction into compact memory cues while preserving raw trajectory pages for intent-driven recall.
• Score: 51.274445160155864
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Long-horizon agentic reasoning requires large language models to act over long interaction histories containing thoughts, tool calls, observations, and partial conclusions. The challenge is not merely that these histories grow long, but that information needed for the current decision may be scattered across distant steps and only become relevant later. Existing approaches address this difficulty by truncating the interaction history, compressing it into shorter surrogates, or retrieving selected parts of it for reuse, but they do not explicitly model how access to past interaction should adapt to the agent's evolving state. We instead cast long-horizon reasoning as a problem of state-adaptive memory. To this end, we propose State-Adaptive Memory~(SAM), a standalone framework that consolidates ongoing interaction into compact memory cues while preserving raw trajectory pages for intent-driven recall. These cues are not treated as replacements for history; rather, they serve as lightweight handles that allow the agent to reconstruct temporally distant information according to its current needs, without retraining the underlying backbone. We further optimize the memory module through expert-guided supervision and reinforcement learning, aligning it with trajectory-level utility. Across BrowseComp, BrowseComp-ZH, WideSearch, and HLE, SAM consistently outperforms strong baselines over diverse agent backbones. Our results suggest that explicit memory modeling provides a simple and effective foundation for long-horizon agentic reasoning.

Related papers

• MRMS: A Multi-Resolution Memory Substrate for Long-Lived AI Agents [2.808851478009314]
  Long-lived AI agents require continuity across interactions.<n> Continuity cannot be obtained by simply extending the prompt window.<n>We propose an architectural memory substrate organized along two axes.
  arXiv  Detail & Related papers  (2026-07-06T02:54:37Z)
• AdMem: Advanced Memory for Task-solving Agents [13.611633561917513]
  Large Language Models (LLMs) show promise as tool-using agents but remain limited in long-horizon tasks that require remembering, organizing, and reusing knowledge.<n>We introduce a unified and automatic memory framework that integrates semantic, episodic, and procedural memory in a bi-level design combining short-term and long-term stores.
  arXiv  Detail & Related papers  (2026-06-05T00:11:57Z)
• Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents [51.30250860677378]
  We propose MRAgent, a framework that combines an associative memory graph with an active reconstruction mechanism.<n>Running on a Cue-Tag-Content graph, our active reconstruction mechanism integrates reasoning directly into memory access.<n>Experiments on the LoCoMo benchmark and LongMemEval benchmark demonstrate significant improvements over strong baselines.
  arXiv  Detail & Related papers  (2026-06-04T11:29:46Z)
• All-Mem: Agentic Lifelong Memory via Dynamic Topology Evolution [20.68235182577703]
  All-Mem maintains a structured memory bank via explicit, non destructive consolidation.<n>Periodically offline, an LLM diagnoser proposes confidence scored topology edits executed with three operators: SPLIT, MERGE, and UPDATE.<n>At query time, typed links enable hop bounded, budgeted expansion from active anchors to archived evidence when needed.
  arXiv  Detail & Related papers  (2026-03-20T03:14:40Z)
• AdaMem: Adaptive User-Centric Memory for Long-Horizon Dialogue Agents [49.63422082885992]
  We propose AdaMem, an adaptive user-centric memory framework for long-horizon dialogue agents.<n>AdaMem organizes dialogue history into working, episodic, persona, and graph memories.<n>We evaluate AdaMem on the LoCoMo and PERSONAMEM benchmarks for long-horizon reasoning and user modeling.
  arXiv  Detail & Related papers  (2026-03-17T13:22:54Z)
• ActMem: Bridging the Gap Between Memory Retrieval and Reasoning in LLM Agents [14.695250837875454]
  We propose a novel actionable memory framework called ActMem.<n>ActMem transforms unstructured dialogue history into a structured causal and semantic graph.<n>It enables agents to deduce implicit constraints and resolve potential conflicts between past states and current intentions.
  arXiv  Detail & Related papers  (2026-02-04T00:54:53Z)
• AMA: Adaptive Memory via Multi-Agent Collaboration [54.490349689939166]
  We propose Adaptive Memory via Multi-Agent Collaboration (AMA), a novel framework that leverages coordinated agents to manage memory across multiple granularities.<n>AMA significantly outperforms state-of-the-art baselines while reducing token consumption by approximately 80% compared to full-context methods.
  arXiv  Detail & Related papers  (2026-01-28T08:09:49Z)
• Memory Matters More: Event-Centric Memory as a Logic Map for Agent Searching and Reasoning [55.251697395358285]
  Large language models (LLMs) are increasingly deployed as intelligent agents that reason, plan, and interact with their environments.<n>To effectively scale to long-horizon scenarios, a key capability for such agents is a memory mechanism that can retain, organize, and retrieve past experiences.<n>We propose CompassMem, an event-centric memory framework inspired by Event Theory.
  arXiv  Detail & Related papers  (2026-01-08T08:44:07Z)
• MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents [84.62985963113245]
  We introduce MEM1, an end-to-end reinforcement learning framework that enables agents to operate with constant memory across long multi-turn tasks.<n>At each turn, MEM1 updates a compact shared internal state that jointly supports memory consolidation and reasoning.<n>We show that MEM1-7B improves performance by 3.5x while reducing memory usage by 3.7x compared to Qwen2.5-14B-Instruct on a 16-objective multi-hop QA task.
  arXiv  Detail & Related papers  (2025-06-18T19:44:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.