Skip to content
AI News HubLIVE
Original source2 min read

MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs

Summary

MemQ introduces a method to enhance LLM agents' episodic memory by applying Q-learning with eligibility traces on a provenance DAG, considering memory dependencies. It achieves top performance across six benchmarks, with significant gains on multi-step tasks.

SourcearXiv AIAuthor: Junwei Liao, Haoting Shi, Ruiwen Zhou, Jiaqian Wang, Shengtao Zhang, Wei Zhang, Weinan Zhang, Ying Wen, Zhiyu Li, Feiyu Xiong, Bo Tang, Muning Wen
MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[2605.08374] MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs

[Submitted on 8 May 2026]

Title:MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs

View a PDF of the paper titled MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs, by Junwei Liao and 11 other authors

View PDF HTML (experimental)

Abstract:Episodic memory allows LLM agents to accumulate and retrieve experience, but current methods treat each memory independently, i.e., evaluating retrieval quality in isolation without accounting for the dependency chains through which memories enable the creation of future memories. We introduce MemQ, which applies TD($\lambda$) eligibility traces to memory Q-values, propagating credit backward through a provenance DAG that records which memories were retrieved when each new memory was created. Credit weight decays as $(\gamma\lambda)^d$ with DAG depth $d$, replacing temporal distance with structural proximity. We formalize the setting as an Exogenous-Context MDP, whose factored transition decouples the exogenous task stream from the endogenous memory store. Across six benchmarks, spanning OS interaction, function calling, code generation, multimodal reasoning, embodied reasoning, and expert-level QA, MemQ achieves the highest success rate on all six in generalization evaluation and runtime learning, with gains largest on multi-step tasks that produce deep and relevant provenance chains (up to +5.7~pp) and smallest on single-step classification (+0.77~pp) where single-step updates already suffice. We further study how $\gamma$ and $\lambda$ interact with the EC-MDP structure, providing principled guidance for parameter selection and future research. Code will be available soon.

Comments: 22 pages, 11 figures (containing 43 individual image panels total)

Subjects:

Artificial Intelligence (cs.AI)

Cite as: arXiv:2605.08374 [cs.AI]

(or arXiv:2605.08374v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2605.08374

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Junwei Liao [view email] [v1] Fri, 8 May 2026 18:30:24 UTC (9,355 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs, by Junwei Liao and 11 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.AI

new | recent | 2026-05

Change to browse by:

cs

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

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? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • MemQ uses TD(λ) eligibility traces to assign credit through provenance DAGs, replacing temporal distance with structural proximity.
  • The method formalizes memory retrieval as an Exogenous-Context MDP, decoupling task stream from memory store.
  • Outperforms baselines on all six benchmarks, with up to +5.7 pp improvement on multi-step tasks.
  • Provides principled guidance for parameter selection via EC-MDP structure.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Agents