InfMem: Learning System-2 Memory Control for Long-Context Agent

• URL: http://arxiv.org/abs/2602.02704v1
• Date: Mon, 02 Feb 2026 19:15:08 GMT
• Title: InfMem: Learning System-2 Memory Control for Long-Context Agent
• Abstract summary: InfMem is a control-centric agent that instantiates System-2-style control via a PreThink-Retrieve-Write protocol.<n>On ultra-long QA benchmarks from 32k to 1M tokens, InfMem consistently outperforms MemAgent across backbones.
• Score: 43.413381779578266
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Reasoning over ultra-long documents requires synthesizing sparse evidence scattered across distant segments under strict memory constraints. While streaming agents enable scalable processing, their passive memory update strategy often fails to preserve low-salience bridging evidence required for multi-hop reasoning. We propose InfMem, a control-centric agent that instantiates System-2-style control via a PreThink-Retrieve-Write protocol. InfMem actively monitors evidence sufficiency, performs targeted in-document retrieval, and applies evidence-aware joint compression to update a bounded memory. To ensure reliable control, we introduce a practical SFT-to-RL training recipe that aligns retrieval, writing, and stopping decisions with end-task correctness. On ultra-long QA benchmarks from 32k to 1M tokens, InfMem consistently outperforms MemAgent across backbones. Specifically, InfMem improves average absolute accuracy by +10.17, +11.84, and +8.23 points on Qwen3-1.7B, Qwen3-4B, and Qwen2.5-7B, respectively, while reducing inference time by $3.9\times$ on average (up to $5.1\times$) via adaptive early stopping.

Related papers

• PARSER: Read in Parallel, Reason in Depth for Long-Context LLM Agents [85.82624962221026]
  We introduce PARSER, which decouples reading from reasoning.<n>A bank of lightweight subagents each bound to a single chunk read the entire document in parallel.<n>On multi-hop QA with contexts ranging from 7K to 896K tokens, PARSER with a 4B backbone outperforms the strongest sequential memory baseline.
  arXiv  Detail & Related papers  (2026-09-06T16:19:01Z)
• Dual-Grained Agent Memory and Shapley Context Attribution for Multimodal Agentic Learner [55.80248281761638]
  multimodal large language models (MLLMs) deliver impressive perception yet still falter on scientific and mathematical reasoning.<n>We propose DG-Mem, a dual-grained agentic memory framework that augments a frozen MLLM with a non- externally stored memory built once from training-time rollouts and consulted read-only at test time.
  arXiv  Detail & Related papers  (2026-08-24T13:56:01Z)
• Stop When Memory Suffices: Evidence-Conditioned Progressive Execution for LLM Agents [5.4775608854724345]
  Existing memory systems either compress and structure histories for efficient access or perform deep research over broader trajectories.<n>This raises a key question: can a memory system achieve strong answer quality while maintaining low online latency?<n>We introduce Router-Mem, an evidence-conditioned progressive execution framework for long-horizon agent memory.
  arXiv  Detail & Related papers  (2026-08-02T14:49:24Z)
• Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents [72.3752981234916]
  Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks.<n>We present Memory as a Controlled Process (MemCon), a framework that models memory operations as a Markov Decision Process.<n>MemCon consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5-20%.
  arXiv  Detail & Related papers  (2026-07-15T08:32:40Z)
• ConvMemory v3: A Validity Context Layer for Conversational Memory via Target-Conditioned Relation Verification [0.0]
  ConvMemory v3 adds a validity context layer that detects and surfaces this update evidence.<n>The core mechanism is a dual-evidence gate that conditions a relation judgment on the specific target proposition.<n>On a synthetic multi-hop validity benchmark the gate reaches 90.12% +/- 1.73 accuracy.
  arXiv  Detail & Related papers  (2026-06-25T08:36:43Z)
• AdaMem: Learning What to Remember for Personalized Long-Horizon LLM Agents [79.03892269184145]
  We argue that what is worth remembering is role-dependent, and propose textbfAdaMem (Adaptive Memory)<n>AdaMem improves QA accuracy by up to textbf+9.0% over the uniform Mem0 baseline while shrinking memory volume by textbf9%.
  arXiv  Detail & Related papers  (2026-06-19T06:35:52Z)
• Feedback-Normalized Developer Memory for Reinforcement-Learning Coding Agents: A Safety-Gated MCP Architecture [0.0]
  This paper presents RL Developer Memory, a local-first, Model Context Protocol (MCP)-native developer-memory architecture for RL coding agents.<n>It treats memory selection as a logged contextual decision process.<n>System is evaluated on a deterministic 200-case benchmark with RL algorithm bugs, hard negatives, review-gated RL/control cases, and low-risk failures.
  arXiv  Detail & Related papers  (2026-05-02T18:37:36Z)
• Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents [0.0]
  coding agents increasingly rely on external memory to reuse prior experience, repair traces, and repository-local operational knowledge.<n>This paper reframes issue-memory use as a selective, risk-sensitive control problem rather than a pure top-k retrieval problem.<n>We introduce RSCB-MC, a risk-sensitive contextual bandit memory controller that decides whether an agent should use no memory, inject the top resolution, summarize multiple candidates, perform high-precision or high-recall retrieval, abstain, or ask for feedback.
  arXiv  Detail & Related papers  (2026-04-30T00:32:53Z)
• D-Mem: A Dual-Process Memory System for LLM Agents [3.5426740232689604]
  We introduce D-Mem, a dual-process memory system.<n>It retains lightweight vector retrieval for routine queries while establishing an exhaustive Full Deliberation module as a high-fidelity fallback.<n>Experiments on the LoCoMo and RealTalk benchmarks using GPT-4o-mini and Qwen3-235B-Instruct demonstrate the efficacy of our approach.
  arXiv  Detail & Related papers  (2026-03-19T08:55:22Z)
• Diagnosing Retrieval vs. Utilization Bottlenecks in LLM Agent Memory [6.493102878734374]
  We analyze how performance differences manifest across write strategies, retrieval methods, and memory utilization behavior.<n>On LoCoMo, retrieval method is the dominant factor.<n>Failure analysis shows that performance breakdowns most often manifest at the retrieval stage rather than at utilization.
  arXiv  Detail & Related papers  (2026-03-02T23:47:23Z)
• Training Large Reasoning Models Efficiently via Progressive Thought Encoding [63.254758972725654]
  Large reasoning models (LRMs) excel on complex problems but face a critical barrier to efficiency.<n>We introduce Progressive Thought, a parameter-efficient fine-tuning method that enables LRMs to reason effectively under fixed-size caches.
  arXiv  Detail & Related papers  (2026-02-18T20:03:38Z)
• Out of the Memory Barrier: A Highly Memory Efficient Training System for LLMs with Million-Token Contexts [68.79341332280062]
  Training Large Language Models (LLMs) on long contexts is severely constrained by prohibitive GPU memory overhead, not training time.<n>We introduce OOMB, a highly memory-efficient training system that directly confronts this barrier.<n>Our approach employs a chunk-recurrent training framework with on-the-fly activation recomputation, which maintains a constant activation memory footprint.
  arXiv  Detail & Related papers  (2026-02-02T13:52:40Z)
• Mem-T: Densifying Rewards for Long-Horizon Memory Agents [23.19373149519922]
  We introduce Mem-T, an autonomous memory agent that interfaces with a lightweight hierarchical memory database to perform dynamic updates and multi-turn retrieval over streaming inputs.<n>We also propose MoT-GRPO, a tree-guided reinforcement learning framework that transforms sparse terminal feedback into dense, step-wise supervision via memory operation tree backpropagation and hindsight credit assignment.
  arXiv  Detail & Related papers  (2026-01-30T14:23:33Z)
• Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents [57.38404718635204]
  Large language model (LLM) agents face fundamental limitations in long-horizon reasoning due to finite context windows.<n>Existing methods typically handle long-term memory (LTM) and short-term memory (STM) as separate components.<n>We propose Agentic Memory (AgeMem), a unified framework that integrates LTM and STM management directly into the agent's policy.
  arXiv  Detail & Related papers  (2026-01-05T08:24:16Z)
• MemR$^3$: Memory Retrieval via Reflective Reasoning for LLM Agents [29.652985606497882]
  We build memory retrieval as an autonomous, accurate, and compatible agent system.<n>MemR$3$ has two core mechanisms: 1) a router that selects among retrieve, reflect, and answer actions to optimize answer quality; 2) a global evidence-gap tracker that explicitly renders the answering process transparent and tracks the evidence collection process.
  arXiv  Detail & Related papers  (2025-12-23T10:49:42Z)
• QwenLong-L1.5: Post-Training Recipe for Long-Context Reasoning and Memory Management [81.24674400474989]
  QwenLong-L1.5 is a model that provides superior longcontext reasoning capabilities through systematic post-training innovations.<n>We develop a systematic framework that generates challenging reasoning tasks requiring multihop sequences over globally distributed evidence.<n>We develop a memory management framework with multi-exploit fusion RL training that seamlessly integrates single-pass reasoning with iterative memory-based processing for tasks exceeding 4M tokens.
  arXiv  Detail & Related papers  (2025-12-15T04:11:11Z)
• Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution [52.76038908826961]
  We propose $textbfReMe$ ($textitRemember Me, Refine Me$) to bridge the gap between static storage and dynamic reasoning.<n>ReMe innovates across the memory lifecycle via three mechanisms: $textitmulti-faceted distillation$, which extracts fine-grained experiences.<n>Experiments on BFCL-V3 and AppWorld demonstrate that ReMe establishes a new state-of-the-art in agent memory system.
  arXiv  Detail & Related papers  (2025-12-11T14:40:01Z)
• MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement Learning [73.27233666920618]
  We propose MemSearcher, an agent workflow that iteratively maintains a compact memory and combines the current turn with it.<n>At each turn, MemSearcher fuses the user's question with the memory to generate reasoning traces, perform search actions, and update memory to retain only information essential for solving the task.<n>We introduce multi-context GRPO, an end-to-end RL framework that jointly optimize reasoning, search strategies, and memory management of MemSearcher Agents.
  arXiv  Detail & Related papers  (2025-11-04T18:27:39Z)
• Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents [33.617262543252494]
  We present ReMemR1, a memory-augmented agent with callback-enhanced memory that allows selective retrieval from the entire memory history.<n>We also propose Reinforcement Learning with Multi-Level Rewards (RLMLR), which combines final-answer rewards with dense, step-level signals that guide effective memory use.
  arXiv  Detail & Related papers  (2025-09-27T01:36:46Z)
• PBFT-Backed Semantic Voting for Multi-Agent Memory Pruning [0.0]
  This paper introduces the Co-Forgetting Protocol to enable synchronized memory pruning in multi-agent systems (MAS)<n>The protocol integrates three key components: (1) context-aware voting, where agents utilize a lightweight DistilBERT model to assess the relevance of memory items based on their content and the current operational context; (2) multi-scale temporal decay functions, which assign diminishing importance to memories based on their age and access frequency across different time horizons; and (3) a Practical Byzantine Fault Tolerance (PBFT)-based consensus mechanism, ensuring that decisions to retain or discard memory items are agreed upon by a qualified and fault-tolerant majority of agents
  arXiv  Detail & Related papers  (2025-06-19T08:28:29Z)
• 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.