Search papers, labs, and topics across Lattice.
Renmin University of China
2
0
4
Bridging the gap between transient experience and persistent capabilities, SPEE enables LLMs to self-improve by evolving their knowledge through a unique experience distillation process.
LEEPS achieves a remarkable 3.7% performance boost in reasoning tasks by intelligently balancing prompt exploitation and exploration, redefining efficiency in RLVR for large language models.
