MemBodied: Recurrent Associative Memory for Vision-Language-Action Models

Hugging Face Daily Papers Papers

Summary

MemBodied introduces a fixed-size episodic memory with associative state and episode anchor to enhance vision-language-action models for history-dependent manipulation tasks, achieving significant performance improvements over baselines.

Vision-Language-Action models provide a strong foundation for general-purpose robot control, yet a vast majority of policies do not preserve and leverage episode-level information beyond the current observation. This limitation is consequential in history-dependent manipulation tasks that depend on information available only in past observations. Retaining past observations in context can aid in recovering this information, but at the significant cost of ever-growing, bloated context and inference latency. We thus introduce MemBodied, a fixed-size episodic memory with two complementary components: an associative state that records interactions across policy calls and an episode anchor that preserves a compact representation of the initial scene as a reference. At each policy call, the model conditions action generation on the current input and the memory components, rather than directly using past observations. Across five evaluated RMBench tasks requiring memory, MemBodied achieves 7.81times the mean success rate of a stateless policy and 2.98times of vanilla recurrent memory, while outperforming the strongest memory-augmented baseline by 1.3times with 10times fewer added parameters. On the fully observable LIBERO-Long suite, it reached 90.6%, a 5.4% improvement over the stateless π_0 policy. These findings support MemBodied as a practical alternative to expanding the policy context for history-dependent manipulation.
Original Article
View Cached Full Text

Cached at: 09/24/26, 03:37 AM

Paper page - MemBodied: Recurrent Associative Memory for Vision-Language-Action Models

Source: https://huggingface.co/papers/2609.28256

Abstract

Vision-Language-Actionmodelsprovideastrongfoundationforgeneral-purposerobotcontrol,yetavastmajorityofpoliciesdonotpreserveandleverageepisode-levelinformationbeyondthecurrentobservation.Thislimitationisconsequentialinhistory-dependentmanipulationtasksthatdependoninformationavailableonlyinpastobservations.Retainingpastobservationsincontextcanaidinrecoveringthisinformation,butatthesignificantcostofever-growing,bloatedcontextandinferencelatency.WethusintroduceMemBodied,afixed-sizeepisodicmemorywithtwocomplementarycomponents:anassociativestatethatrecordsinteractionsacrosspolicycallsandanepisodeanchorthatpreservesacompactrepresentationoftheinitialsceneasareference.Ateachpolicycall,themodelconditionsactiongenerationonthecurrentinputandthememorycomponents,ratherthandirectlyusingpastobservations.AcrossfiveevaluatedRMBenchtasksrequiringmemory,MemBodiedachieves7.81timesthemeansuccessrateofastatelesspolicyand2.98timesofvanillarecurrentmemory,whileoutperformingthestrongestmemory-augmentedbaselineby1.3timeswith10timesfeweraddedparameters.OnthefullyobservableLIBERO-Longsuite,itreached90.6%,a5.4%improvementoverthestatelessπ_0policy.ThesefindingssupportMemBodiedasapracticalalternativetoexpandingthepolicycontextforhistory-dependentmanipulation.

View arXiv pageView PDFProject pageGitHub1Add to collection

Get this paper in your agent:

hf papers read 2609\.28256

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/2609.28256 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.28256 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.28256 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