1
Lightweight Latent Reasoning for Narrative Tasks
轻量级潜推理技术利用强化学习优化LLM生成推理轨迹,提升叙事类任务表现
arXiv:2512.02240v2 Announce Type: replace Abstract: Large language models (LLMs) tackle complex tasks by generating long chains of thought or "reasoni…
轻量级潜推理技术利用强化学习优化LLM生成推理轨迹,提升叙事类任务表现
arXiv:2512.02240v2 Announce Type: replace Abstract: Large language models (LLMs) tackle complex tasks by generating long chains of thought or "reasoni…
大模型推理时的“内心独白”可能被窃取,一篇揭示LLM推理轨迹暴露风险的研究
arXiv:2606.00642v1 Announce Type: new Abstract: Reasoning traces have become a valuable form of learning signals for improving and transferring the ca…
新方法将LLM推理轨迹分割为操作符,提升可解释性与效率
arXiv:2605.29192v1 Announce Type: new Abstract: Chain-of-thought traces from large reasoning models can span tens of thousands of tokens, yet we lack …