2026-10-14 · America/Los_Angeles · 社区动态 · #7

Reasoning models burn tokens they don't need: a training-free contrastive decoding method shortens the chain of thought without retraining

Reasoning models burn tokens they don't need: a training-free contrastive decoding method shortens the chain of thought without retraining

热度 50.6 / 100；排名与评分保留该期记录。