Search papers, labs, and topics across Lattice.
Affiliation:
1
0
3
0
Turn-level Multiscale Density Ratio Estimation (tlm-DRE), which assigns different weights on corresponding turns and proposes asymmetric token-level training based on the positive-negative space gaps across multiple turns of tasks, enables LLMs to perform robustly in multi-turn reasoning tasks with both in-domain and out-of-domain conditions.
