Preprint
Jul 2026
Procedural Memory Distillation: Online Reflection for Self-Improving Language Models
Procedural Memory Distillation is proposed, which converts crossepisode signals into reusable procedural memory and distills it into the policy's weights during training, yielding a memory-free model at inference.
Ye Liu, Srijan Bansal, Bo Pang et al.
· 2 citations