AURA: Action-Gated Memory for Robot Policies at Constant VRAM

In three linesAURA-Mem introduces a constant-size recurrent memory (4,224 bytes) for robot policies, with a learned gate that writes only when observations would change the next action. On LIBERO-Long with OpenVLA-OFT 7B, it matches baseline policy (0.233 success) while reducing memory writes by 7× and VRAM by 6,061× versus KV-cache.

Summary generated by Claude — human-verified