Search papers, labs, and topics across Lattice.
Baidu Inc.
2
0
3
LLMs can slash over 80% of their chain-of-thought tokens with a minor accuracy boost, thanks to a new RL-based method that targets the "Minimal Sufficient Length" of reasoning.
LLMs can escape the trap of converging on popular but incorrect answers in unsupervised RLVR by temporarily "unlearning" and exploring diverse response options.
