Search papers, labs, and topics across Lattice.
Shanghai Artificial Intelligence Laboratory, University of Science and Technology
3
0
6
Transforming textual skills into adaptable parameters at test time boosts LLM performance by over 6 points in complex software engineering tasks.
Evolving generative models in residual space reveals a powerful balance between local refinement and global exploration, enhancing data editing capabilities.
Offline RL can now refine trajectories conservatively without risking extrapolation, thanks to a novel framework that leverages local preference pairs for targeted improvements.
