Search papers, labs, and topics across Lattice.
Zhejiang University
2
0
5
DeferMem's RL-trained memory distiller lets LLMs cherry-pick and rewrite scattered information into query-specific evidence, slashing memory costs and boosting QA accuracy.
LLMs can now actively perceive and react to anomalies during scientific simulations, leading to more reliable and accurate results in complex engineering and modeling tasks.
