EMCES: Episodic Memory Guides RL Data Generation
jiqizhixin · x · 2026-07-10
Researchers from Zhejiang Sci-Tech University and Nanjing University introduced the EMCES system, which uses episodic memory to guide diffusion models in generating high-quality synthetic training data for reinforcement learning, prioritizing the most valuable past experiences.
Experiments show that this method outperforms existing approaches across multiple environments, effectively boosting RL algorithm performance. Additionally, the research leveraged hashing tricks to reduce storage space by 8000x and time by 25.5x without any loss in quality.
More from Research
- LLM leaderboards are now often measuring the harness too, Gary Marcus warns — GaryMarcus · 2026-07-22
- New paper defines self-state attacks, showing OS defenses leave four agent-memory cases indistinguishable — Justgototheeffinmoon · 2026-07-22
- Krea 2 users recommend a two-pass Clownshark sampler setup for sharper image details — listopalafoto · 2026-07-22
- Animation shows how an MLP’s first-layer weights change while learning MNIST — CatAstro_Piyush · 2026-07-22
- Project APE finds verifier reliability drops when papers contain multiple errors — soumitrashukla9 · 2026-07-22
- Project APE says verifier costs fell about 90x in a year as Chinese open models lead — soumitrashukla9 · 2026-07-22