cs.LGNov 17, 2026

Perception into Imagination: Executable Memory for World Models

Authors: Xianyao Li, Ruitong Tian, Rui Min, Fang Xu, Jing Du

Abstract

Physical agents must use what they previously saw or touched after visibility or contact is lost. Their actions change the implications of these remembered facts. We introduce executable perceptual memory (EPM), a world-model framework that advances explicit perceptual state alongside neural latent dynamics during imagination. Its evolving read-out informs reward prediction, value estimation and policy learning. We align reward inputs with the transitions that produced them and supervise the reward head on open-loop rollouts of recorded actions. On Memory Maze 13×1313\times13, EPM reaches 30.37 return at 1M interactions, compared with 23.37 when the same evolving read-out informs only policy and value on a common perceptual front end: a gain of 7.00 [3.80, 10.21]. The complete method also improves control on 9×99\times9 and in MiniWorld. Probes of remembered, out-of-view arrivals show stronger reward anticipation in both visual implementations. Concentration, tactile recovery and real tactile frames demonstrate complementary roles of remembered facts in imagined-policy training, motion-dependent decisions and sensory-content retention. EPM connects persistent perception to the predicted consequences and learning signals of a world model for Physical AI.

Figures & tables

Appendix figures & tables13 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. MemWM: Memory-Augmented Text-Based World Model

    Aug 7, 2026Yujun Wang, Tao Zhang, Jinhe Bi +9LLM World ModelsWorld Model Learning

  2. Mem-World: Memory-Augmented Action-Conditioned World Models for Persistent Robot Manipulation

    Jun 17, 2026Zirui Zheng, Jiaqian Yu, Xiongfeng Peng +7Long-Horizon Robotic ManipulationWorld Models for Robotics

  3. Dream to Remember: Continual Steering of Generalist Robot Policies with World Models

    Oct 9, 2026Bahey Tharwat, Iman Nematollahi, Marvin Alles +3Continual Robot LearningRobotic RL