Search papers, labs, and topics across Lattice.
Fudan University Renmin University of China
5
0
8
External agent scaffolding can be discarded after training by distilling evolved decision-making procedures directly into a model's native reasoning traces.
Verification of coding agent outputs is now the bottleneck, not generation, and targeted design can significantly enhance performance while curbing reward hacking.
Learned critics in RLHF can actually *increase* variance and hurt performance in sparse-reward settings, but a simple explained variance metric can tell you when to ditch the critic and get better results.
Multi-turn reinforcement learning gets a boost: weighting trajectories by semantic similarity dramatically improves baseline estimation and agent performance in long-document visual QA.
Multi-hop data synthesis using HopChain boosts VLM performance across a wide range of tasks, with gains of over 50 points in accuracy for ultra-long-context reasoning.
