Search papers, labs, and topics across Lattice.
Zhejiang University
5
0
7
6
HSS-Synth generates 237k high-quality instruction samples that not only outperform existing baselines but also redefine data synthesis for the humanities and social sciences.
BridgeAlign achieves unprecedented alignment in humanities and social sciences, enabling LLMs to excel in both human-preference and knowledge-based tasks simultaneously.
TRACE redefines rollout budget allocation by treating each turn in a multi-turn interaction as a unique node, leading to improved reward contrast and policy performance.
ISPO reduces critical reasoning failures in RLVR by transforming reward structures, leading to superior performance on complex reasoning tasks.
Synthetic data that looks good can still tank your model's performance – Optimsyn uses influence functions to find the *actually* useful synthetic examples and optimize your generation rubrics.
