[2510.12635] Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks

-->

Computer Science > Artificial Intelligence

arXiv:2510.12635 (cs)

[Submitted on 14 Oct 2025 (v1), last revised 7 May 2026 (this version, v3)]

Title:Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks

Authors:Yuxiang Zhang, Jiangming Shu, Ye Ma, Xueyuan Lin, Shangxi Wu, Jitao Sang<br>View a PDF of the paper titled Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks, by Yuxiang Zhang and 5 other authors

View PDF<br>HTML (experimental)

Abstract:Long-context Large Language Models, despite their expanded capacity, require careful working memory management to mitigate attention dilution during long-horizon tasks. Yet existing approaches rely on external mechanisms that lack awareness of the agent's reasoning state, leading to suboptimal decisions. We propose Memory-as-Action (MemAct), a framework that treats working memory management as learnable policy actions. By formulating context management as in-place editing operations (deletion, insertion), MemAct enables joint optimization of information retention and task performance through end-to-end reinforcement learning. To address the computational challenges of dynamic context updates, we introduce Dynamic Context Policy Optimization, which restores training efficiency without compromising reasoning integrity. Experiments show that MemAct-RL-14B matches the accuracy of models $16\times$ larger while reducing average context length by 51\%, with learned strategies that adapt to model capabilities and generalize across task complexities.

Subjects:

Artificial Intelligence (cs.AI)

Cite as:<br>arXiv:2510.12635 [cs.AI]

(or<br>arXiv:2510.12635v3 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2510.12635

Focus to learn more

arXiv-issued DOI via DataCite

Submission history<br>From: Yuxiang Zhang [view email]<br>[v1]<br>Tue, 14 Oct 2025 15:29:57 UTC (327 KB)

[v2]<br>Sat, 10 Jan 2026 01:44:56 UTC (374 KB)

[v3]<br>Thu, 7 May 2026 13:18:53 UTC (371 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks, by Yuxiang Zhang and 5 other authors<br>View PDF<br>HTML (experimental)<br>TeX Source

view license

Current browse context:

cs.AI

next >

new<br>recent<br>| 2025-10

Change to browse by:

cs

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Which authors of this paper are endorsers? |<br>Disable MathJax (What is MathJax?)