Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents

Hugging Face Daily Papers Papers
memory agent long-horizon proactive-memory decision-making plug-and-play open-source

Summary

This paper introduces a proactive memory agent that runs alongside an action agent to prevent behavioral state decay in long-horizon tasks, achieving significant improvements on Terminal-Bench2.0 and τ^2-Bench. The authors also train Qwen3.5-27B using SFT and GRPO as an early step toward open-weight memory policies.

In long-horizon tasks, decision-relevant state is often scattered across an expanding trajectory, while the action agent must surface it and act. As trajectories grow, task requirements, environment facts, prior attempts, diagnoses, and open subgoals can be buried in the context window or pushed beyond it, failing to influence decisions when needed. We call this failure mode "behavioral state decay". We study memory as an active intervention mechanism rather than passive retrieval. A separate memory agent runs alongside an unmodified action agent, updating a structured memory bank from the recent trajectory and deciding whether to inject a memory-grounded reminder or remain silent. The module is plug-and-play with frontier action agents and existing agent harnesses. Across Terminal-Bench 2.0 and τ^2-Bench, it improves pass@1 for both weaker and stronger action agents, with gains of +8.3 pp on Terminal-Bench and +6.8 pp on τ^2-Bench. Ablations show that selective intervention outperforms passive bank exposure, always-on injection, advisor-only guidance, and general retrieval. As an early step toward open-weight memory policies, we train Qwen3.5-27B on SETA using SFT and GRPO, improving validation reward and achieving partial transfer to Terminal-Bench.
Original Article
View Cached Full Text

Cached at: 07/10/26, 10:12 PM

Paper page - Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents

Source: https://huggingface.co/papers/2607.08716

Abstract

Inlong-horizontasks,decision-relevantstateisoftenscatteredacrossanexpandingtrajectory,whiletheactionagentmustsurfaceitandact.Astrajectoriesgrow,taskrequirements,environmentfacts,priorattempts,diagnoses,andopensubgoalscanbeburiedinthecontextwindoworpushedbeyondit,failingtoinfluencedecisionswhenneeded.Wecallthisfailuremode“behavioralstatedecay“.Westudymemoryasanactiveinterventionmechanismratherthanpassiveretrieval.Aseparatememoryagentrunsalongsideanunmodifiedactionagent,updatingastructuredmemorybankfromtherecenttrajectoryanddecidingwhethertoinjectamemory-groundedreminderorremainsilent.Themoduleisplug-and-playwithfrontieractionagentsandexistingagentharnesses.AcrossTerminal-Bench2.0andτ^2-Bench,itimprovespass@1forbothweakerandstrongeractionagents,withgainsof+8.3pponTerminal-Benchand+6.8pponτ^2-Bench.Ablationsshowthatselectiveinterventionoutperformspassivebankexposure,always-oninjection,advisor-onlyguidance,andgeneralretrieval.Asanearlysteptowardopen-weightmemorypolicies,wetrainQwen3.5-27BonSETAusingSFTandGRPO,improvingvalidationrewardandachievingpartialtransfertoTerminal-Bench.

View arXiv pageView PDFGitHub0Add to collection

Get this paper in your agent:

hf papers read 2607\.08716

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.08716 in a model README.md to link it from this page.

Datasets citing this paper1

#### cy0307/awesome-loop-engineering Viewer• Updatedabout 5 hours ago • 430 • 1.2k • 2

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.08716 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Proactive Memory for Long-Horizon Agents (16 minute read)

TLDR AI

This paper introduces a proactive memory agent that operates alongside a standard action agent to selectively inject memory-grounded reminders during long-horizon tasks, mitigating behavioral state decay. Experiments on Terminal-Bench and τ²-Bench show significant improvements in pass@1, and the approach is demonstrated with both weak and strong action agents.

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

Hugging Face Daily Papers

This paper proposes SAM, a state-adaptive memory framework that dynamically manages interaction histories for long-horizon agentic reasoning, enabling intent-driven recall without retraining the backbone model. It outperforms strong baselines across multiple benchmarks like BrowseComp and HLE.

PREPING: Building Agent Memory without Tasks

Hugging Face Daily Papers

Presents PREPING, a framework for constructing agent memory before any task-specific experience using proposer-guided synthetic practice, achieving competitive performance with reduced deployment costs.