Guardado en:
Detalles Bibliográficos
Autores principales: Sridhar, Ajay, Pan, Jennifer, Sharma, Satvik, Finn, Chelsea
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2510.20328
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909865076588544
author Sridhar, Ajay
Pan, Jennifer
Sharma, Satvik
Finn, Chelsea
author_facet Sridhar, Ajay
Pan, Jennifer
Sharma, Satvik
Finn, Chelsea
contents Humans routinely rely on memory to perform tasks, yet most robot policies lack this capability; our goal is to endow robot policies with the same ability. Naively conditioning on long observation histories is computationally expensive and brittle under covariate shift, while indiscriminate subsampling of history leads to irrelevant or redundant information. We propose a hierarchical policy framework, where the high-level policy is trained to select and track previous relevant keyframes from its experience. The high-level policy uses selected keyframes and the most recent frames when generating text instructions for a low-level policy to execute. This design is compatible with existing vision-language-action (VLA) models and enables the system to efficiently reason over long-horizon dependencies. In our experiments, we finetune Qwen2.5-VL-7B-Instruct and $π_{0.5}$ as the high-level and low-level policies respectively, using demonstrations supplemented with minimal language annotations. Our approach, MemER, outperforms prior methods on three real-world long-horizon robotic manipulation tasks that require minutes of memory. Videos and code can be found at https://jen-pan.github.io/memer/.
format Preprint
id arxiv_https___arxiv_org_abs_2510_20328
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MemER: Scaling Up Memory for Robot Control via Experience Retrieval
Sridhar, Ajay
Pan, Jennifer
Sharma, Satvik
Finn, Chelsea
Robotics
Artificial Intelligence
Machine Learning
Humans routinely rely on memory to perform tasks, yet most robot policies lack this capability; our goal is to endow robot policies with the same ability. Naively conditioning on long observation histories is computationally expensive and brittle under covariate shift, while indiscriminate subsampling of history leads to irrelevant or redundant information. We propose a hierarchical policy framework, where the high-level policy is trained to select and track previous relevant keyframes from its experience. The high-level policy uses selected keyframes and the most recent frames when generating text instructions for a low-level policy to execute. This design is compatible with existing vision-language-action (VLA) models and enables the system to efficiently reason over long-horizon dependencies. In our experiments, we finetune Qwen2.5-VL-7B-Instruct and $π_{0.5}$ as the high-level and low-level policies respectively, using demonstrations supplemented with minimal language annotations. Our approach, MemER, outperforms prior methods on three real-world long-horizon robotic manipulation tasks that require minutes of memory. Videos and code can be found at https://jen-pan.github.io/memer/.
title MemER: Scaling Up Memory for Robot Control via Experience Retrieval
topic Robotics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2510.20328