cola-test-time-depth-adaptation-llm

【开发与工程】Dynamically adapt LLM depth per input at test time by skipping, repeating, or reordering layers using MCTS search, correcting 60% of initially wrong predictions and processing 75% of correct predictions with shorter architectures without retraining.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准