Search papers, labs, and topics across Lattice.
Affiliation:
3
0
6
FlexComp is proposed, a method-agnostic framework that decouples the ratio from both training and deployment and matches separately trained fixed-ratio specialists with minimal degradation across ICAE, 500xCompressor, and SAC on MRQA.
A frozen instruct model can reshape a student's reasoning policy, enabling RL refinement that surpasses traditional methods without costly fine-tuning.
Dynamic data mixing via loss trajectories boosts performance across tasks while using just 25% of the proxy compute budget.
