Search papers, labs, and topics across Lattice.
The paper introduces Gated Memory Policy (GMP), a visuomotor policy that selectively incorporates historical context using a learned memory gate and cross-attention module to address the performance degradation observed when naively extending observation histories in robotic manipulation tasks. GMP learns when to recall memory via the gate and what to recall via cross-attention, while also injecting diffusion noise into historical actions to improve robustness. Experiments on a new non-Markovian benchmark, MemMimic, demonstrate a 30.1% improvement over long-history baselines, while maintaining performance on Markovian tasks.
Simply feeding more history to visuomotor policies hurts performance; GMP solves this by learning when and what to remember, boosting success rates by 30% on memory-intensive robotic tasks.
Robotic manipulation tasks exhibit varying memory requirements, ranging from Markovian tasks that require no memory to non-Markovian tasks that depend on historical information spanning single or multiple interaction trials. Surprisingly, simply extending observation histories of a visuomotor policy often leads to a significant performance drop due to distribution shift and overfitting. To address these issues, we propose Gated Memory Policy (GMP), a visuomotor policy that learns both when to recall memory and what to recall. To learn when to recall memory, GMP employs a learned memory gate mechanism that selectively activates history context only when necessary, improving robustness and reactivity. To learn what to recall efficiently, GMP introduces a lightweight cross-attention module that constructs effective latent memory representations. To further enhance robustness, GMP injects diffusion noise into historical actions, mitigating sensitivity to noisy or inaccurate histories during both training and inference. On our proposed non-Markovian benchmark MemMimic, GMP achieves a 30.1% average success rate improvement over long-history baselines, while maintaining competitive performance on Markovian tasks in RoboMimic. All code, data and in-the-wild deployment instructions are available on our project website https://gated-memory-policy.github.io/.
