按年被提到
20252025 · 2 次2026
大模型与语言模型›大模型推理与思维链·arXiv:2501.04682 · cs.AI

Towards System 2 Reasoning in LLMs: Learning How to Think With Meta Chain-of-Thought

Violet Xiang, Charlie Snell, Kanishk Gandhi, Alon Albalak, Anikait Singh, Chase Blagden, Duy Phung, Rafael Rafailov, Nathan Lile, Dakota Mahan, Louis Castricato, Jan-Philipp Franken 等 14 人 · 2025
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 3 人首次 2025-01-09 · 最近 2026-04-04
3不同的人
2活跃月份
4讨论串
37被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We propose a novel framework, Meta Chain-of-Thought (Meta-CoT), which extends traditional Chain-of-Thought (CoT) by explicitly modeling the underlying reasoning required to arrive at a particular CoT. We present empirical evidence from state-of-the-art models exhibiting behaviors consistent with in-context search, and explore methods for producing Meta-CoT via process supervision, synthetic data generation, and search algorithms. Finally, we outline a concrete pipeline for training a model to produce Meta-CoTs, incorporating instruction tuning with linearized search traces and reinforcement learning post-training. Finally, we discuss open research questions, including scaling laws, verifier roles, and the potential for discovering novel reasoning algorithms. This work provides a theoretical and practical roadmap to enable Meta-CoT in LLMs, paving the way for more powerful and human-like reasoning in artificial intelligence.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论chunpaiyang · 2026-04

“AI lies. The trick is making it double-check itself before it ships. I made a skill that forces AI to think more. Burns more tokens but improve accuracy. I suggest you use Meta CoT(Chain-of-Thought) https://arxiv.org/abs/2501.04682 to build a think skill to avoid this problem.”

A broken auto-live poller, and what perceived urgency does to Claude Code原评论 ↗
评论chunpaiyang · 2026-04

“I'm a software engineer building an app as a side project. I quickly realized AI bullshits a lot. But you know, engineers bullshit each other all the time too. The difference is we have a way to verify it - logical chains. You have to build an argument that holds up before anayone buys in it. So I though, can I make AI build its own logical chain ? Let …”

LLMs audit code from the same blind spot they wrote it from. Here's the fix原评论 ↗
被提交为 story · 2 次