1
Model soups need only one ingredient
只用一种成分简化模型汤方法，降低集成复杂度的高手秘诀
arXiv:2602.09689v2 Announce Type: replace Abstract: Fine-tuning large pre-trained models on a target distribution often improves in-distribution (ID) …
2
Extrapolative Weight Averaging Reveals Correctness-Efficiency Frontiers in Code RL
外推权重平均法揭示了代码强化学习中正确性与效率的权衡前沿，为模型调优提供新视角。
arXiv:2605.28751v1 Announce Type: cross Abstract: Linear interpolation between fine-tuned checkpoints has been shown to trace the Pareto front between…