AI 与机器学习2025ml-site.cdn-apple.com
思维的错觉：推理模型的能力与局限
The Illusion of Thinking: Strengths and limitations of reasoning models
进阶 · 约 50 分钟读完前置知识：大语言模型基础、推理模型（o1/R1 类）概念、基准测试与评测方法
导读
这篇苹果的论文研究了推理模型（LRM，如 o1/R1 类模型）在 puzzle 类任务（汉诺塔、跳棋、过河、积木世界等）上的真实推理能力。结论是这些模型在复杂度升高后推理能力崩溃，且其“思考过程”可能只是表面上的推理表演，而非真正的规划能力。
HN 关注点：是因为它直接挑战了“推理模型接近 AGI”的流行叙事，讨论焦点在于实验设置是否公平（如是否允许模型写代码）以及语言流畅性是否被误当作智能的证据。
HN 评论精选
这几条评论里：支持 2 质疑 1 补充 1
支持讽刺 AI 炒作周期重演
还记得大家都在喊“AGI 就在眼前”吗？有趣的是 Gartner 炒作周期曲线对这类事情描述得多么准确。
actinium226查看英文原文
Man, remember when everyone was like 'AGI just around the corner!' Funny how well the Gartner hype cycle captures these sorts of things
质疑质疑实验设置：不让模型写代码不公平
论文测试的所有环境（汉诺塔、跳棋、过河、积木世界）如果允许 LLM 写代码，任何模型都能完美解决。我觉得这和“LLM 不会做 20 位数乘法”没什么区别——顺便说，大多数人类也不会，我自己用纸笔试过，总是出错。
thomasahle查看英文原文
All the environments the test (Tower of Hanoi, Checkers Jumping, River Crossing, Block World) could easily be solved perfectly by any of the LLMs if the authors had allowed it to write code. I don't really see how this is different from "LLMs can't multiply 20 digit numbers"--which btw, most humans can't either. I tried it once (using pen and paper) and consistently made errors somewhere.
补充推理轨迹看似人类但内部机制陌生，需要更多此类研究
我们对 LLM 能力感到困惑的一个原因是它们使用语言。看推理轨迹时 token 看起来像人类，但实际发生的过程对我们来说非常陌生，正如《Biology of Large Language Models》和《Safety Alignment Should Be Made More Than Just a Few Tokens Deep》所展示的。我很难看清这项技术能做什么、不能做什么，尤其是在设计系统时。我认为这类研究很棒，我们需要投入更多精力去理解如何使用这些 token 以及如何基于它们构建系统。
jackdoe查看英文原文
I think one of the reason we are confused about what LLMs can do is because they use language. And we look at the "reasoning traces" and the tokens there look human, but what is actually happening is very alien to us, as shown by "Biology of Large Language Models"[1] and "Safety Alignment Should Be Made More Than Just a Few Tokens Deep"[2] I am struggling a lot to see what the tech can and can not do, particularly designing systems with them, and how to build systems where the whole is bigger than the sum of its parts. And I think this is because I am constantly confused by their capabilities, despite understanding their machinery and how they work, their use of language just seems like magic. I even wrote https://punkx.org/jackdoe/language.html just to remind myself how to think about it. I think this kind of research is amazing and we have to spend tremendous more effort into understanding how to use the tokens and how to build with them. [1]: https://transformer-circuits.pub/2025/attribution-graphs/biology.html [2]: https://arxiv.org/pdf/2406.05946
支持语言流畅性被误当作智能的代理指标
人类语言作为认知工具远不完美，但仍有用，因为它不是基础性的，只是高层抽象层。我坚信人类语言太弱（模糊、不一致、表达力不足），无法替代与世界的交互来构建强认知。我们之所以容易被 LLM/LRM 的结果愚弄，是因为我们习惯把语言流畅性和知识检索当作衡量同伴智力的代理指标。
stephc_int13查看英文原文
Human language is far from perfect as a cognitive tool but still serves us well because it is not foundational. We use it both for communication and some reasoning/planning as a high level layer. I strongly believe that human language is too weak (vague, inconsistent, not expressive enough etc.) to replace interactions with the world as a basis to build strong cognition. We're easily fooled by the results of LLM/LRM models because we typically use language fluency and knowledge retrieval as a proxy benchmark for intelligence among our peers.
相关资源
从这篇论文所有 HN 讨论帖的全部评论里整理出来的链接。
历次讨论
- 2025Hacker News 讨论帖488 分 · 270 评
同领域还值得读
DeepSeek-R1：通过强化学习激励 LLM 的推理能力
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL
研究如何用强化学习激励大语言模型的推理能力，提出 DeepSeek-R1，并探讨 RL 训练对模型推理表现的影响。
AI 与机器学习2025 · 1,351 分 · 1,056 评 · 引用 30进阶 · 约 80 分钟编辑精选1-bit LLM 时代：面向低成本计算的三值参数
The Era of 1-bit LLMs: ternary parameters for cost-effective computing
提出用三值参数 {-1, 0, 1} 替代 LLM 中的浮点权重，即所谓“1-bit LLM”。
AI 与机器学习2024 · 1,040 分 · 447 评 · 引用 19进阶 · 约 45 分钟编辑精选DeepSeek-V3.2：推进开放大语言模型的前沿
DeepSeek-v3.2: Pushing the frontier of open large language models
这是 DeepSeek-V3.2 的技术报告，延续了 DeepSeek 开源权重模型的路线，目标是让开放模型继续逼近甚至追平前沿闭源模型。
AI 与机器学习2025 · 982 分 · 465 评进阶 · 约 90 分钟编辑精选通过自我对弈与通用强化学习掌握国际象棋和将棋（AlphaZero）
Mastering Chess and Shogi by Self-Play with General Reinforcement Learning
DeepMind 的 AlphaZero 论文：用通用强化学习算法（自我博弈 + MCTS + 神经网络），从零开始仅数小时训练便在国际象棋和将棋上击败了世界顶级引擎 Stock…
AI 与机器学习2017 · 539 分 · 270 评 · 引用 13进阶 · 约 90 分钟编辑精选用强化学习发现更快的矩阵乘法算法
Discovering faster matrix multiplication algorithms with reinforcement learning
论文提出 AlphaTensor，用强化学习搜索矩阵乘法算法。
AI 与机器学习2022 · 518 分 · 111 评专业 · 约 90 分钟编辑精选