Recursive Experiential-Working Memory Evolution for Long-Horizon Agent Harnesses

Hugging Face Daily Papers Papers
recursive-memory long-horizon-agents ai-research agent-architecture memory-evolution validation-gated-updates

Summary

The paper introduces Recuris, a recursive memory architecture that improves long-horizon agent success by tracking progress and guiding skill selection through localized, validation-gated updates.

Recursive self-improvement (RSI) remains hard in long-horizon tasks, where growing histories obscure the task state and misalign skill invocation. We introduce Recuris, a recursive Experiential-Working Memory architecture for long-horizon agent harnesses, in which Working Memory tracks task progress and guides skill selection from Experiential Memory, grounding skill use in current needs rather than the full history. This coupling also turns execution into structured evidence that localizes failures to specific memory components. Across tasks, a fixed Meta-Agent turns that evidence into localized, validation-gated updates to Skill Memory that reshape execution and yield new evidence, forming a bounded recursive memory-evolution loop. Across four long-horizon benchmarks and ten models, Recuris improves task success in 35 of the 37 completed model-benchmark pairs, carrying frontier models to SOTA-level task success: on tau-bench it adds +17.8 points to GPT-5.6 Sol and +15.6 to Claude Opus 5, taking Opus 5 to 87.9%, and +16.6/+13.5 points on Qwen3.6-27B/35B on SkillFlow. The advantage widens as the interaction horizon grows, to +32.2 points on the longest tasks, and common long-horizon failures fall by up to 80%. These results position recursively evolving memory as a scalable foundation for RSI, enabling agents to continuously transform accumulated experience into increasingly effective long-horizon behavior. Code: https://github.com/Gen-Verse/Recuris
Original Article
View Cached Full Text

Cached at: 08/26/26, 03:16 AM

Paper page - Recursive Experiential-Working Memory Evolution for Long-Horizon Agent Harnesses

Source: https://huggingface.co/papers/2608.24876

Abstract

Recuris introduces a recursive memory architecture that tracks progress and guides skill selection to improve long-horizon agent success through localized, validation-gated updates.

Recursive self-improvement(RSI) remains hard in long-horizon tasks, where growing histories obscure the task state and misalign skill invocation. We introduce Recuris, a recursiveExperiential-Working Memoryarchitecture forlong-horizon agent harnesses, in whichWorking Memorytracks task progress and guides skill selection fromExperiential Memory, grounding skill use in current needs rather than the full history. This coupling also turns execution into structured evidence that localizes failures to specific memory components. Across tasks, a fixedMeta-Agentturns that evidence into localized,validation-gated updatestoSkill Memorythat reshape execution and yield new evidence, forming a bounded recursive memory-evolution loop. Across four long-horizon benchmarks and ten models, Recuris improves task success in 35 of the 37 completed model-benchmark pairs, carrying frontier models to SOTA-level task success: on tau-bench it adds +17.8 points to GPT-5.6 Sol and +15.6 to Claude Opus 5, taking Opus 5 to 87.9%, and +16.6/+13.5 points on Qwen3.6-27B/35B on SkillFlow. The advantage widens as the interaction horizon grows, to +32.2 points on the longest tasks, and common long-horizon failures fall by up to 80%. These results position recursively evolving memory as a scalable foundation for RSI, enabling agents to continuously transform accumulated experience into increasingly effective long-horizon behavior. Code: https://github.com/Gen-Verse/Recuris

View arXiv pageView PDFProject pageGitHub4Add to collection

Get this paper in your agent:

hf papers read 2608\.24876

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/2608.24876 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.24876 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.24876 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.

ERSkill: Evolving for Skill-Guided Adaptive Memory Retrieval

arXiv cs.CL

Introduces ERSkill, a retrieval-centric framework for self-evolving, skill-guided adaptive memory access in LLM agents. It co-evolves retrieval skills and a routing policy, substantially outperforming strong baselines across agent memory benchmarks.

Harnessing agent memory to build lifelong AI partners for materials scientists

arXiv cs.AI

This paper proposes a self-evolving memory framework for lifelong AI partners in materials science, storing scientific experience as facts and skills to improve agent performance across models. Evaluations show memory nearly doubles GPT-5.2 task success in materials tool-use questions and reduces repeated errors in simulations.

Self-Improving Memory for Agents (6 minute read)

TLDR AI

Perplexity Brain is a memory system that builds a persistent context graph across tasks, projects, decisions, files, and sources, enabling agents to start with relevant context instead of from scratch, improving answer correctness and reducing task costs.