Divide-and-Remember: Recursive Action-Relevant Memory for Long-Horizon VLA Policies

Name: Divide-and-Remember: Recursive Action-Relevant Memory for Long-Horizon VLA Policies
Uploaded: 2026-10-05T11:35:12.164619+00:00
Duration: 569 s
Description: Vision-language-action (VLA) models struggle with long tasks because they cannot remember past information effectively. This work proposes a memory function that maximizes mutual information between actions and memory given observations. The Divide-and-Remember (D&R) method implements this optimal memory recursively, allowing VLA models to retain relevant past context efficiently.

Vision-language-action (VLA) models struggle with long tasks because they cannot remember past information effectively. This work proposes a memory function that maximizes mutual information between actions and memory given observations. The Divide-and-Remember (D&R) method implements this optimal memory recursively, allowing VLA models to retain relevant past context efficiently.

Open the full episode