Cot_compress - 标签 - vllbc02's blogs

Cot_compress

2025

Less is More Tokens：Efficient Math Reasoning via Difficulty-Aware Chain-of-Thought Distillation 09-09

ThinkDial：An Open Recipe for Controlling Reasoning Effort in Large Language Models 08-29

Think in Blocks：Adaptive Reasoning from Direct Response to Deep Reasoning 08-23

Pruning Long Chain-of-Thought of Large Reasoning Models via Small-Scale Preference Optimization 08-18

Promoting Efficient Reasoning with Verifiable Stepwise Reward 08-18

Sample More to Think Less：Group Filtered Policy Optimization for Concise Reasoning 08-15

TRAIN LONG, THINK SHORT：CURRICULUM LEARNING FOR EFFICIENT REASONING 08-14

LAPO：INTERNALIZING REASONING EFFICIENCY VIA LENGTH-ADAPTIVE POLICY OPTIMIZATION 08-04