用“推理能量”量化思维链每步开销，为LLM思考效率提供新视角
arXiv:2607.28674v1 Announce Type: new Abstract: Understanding how computational effort is allocated across individual chain-of-thought (CoT) reasoning…
大模型坐标基准的隐藏陷阱：RMSNorm下残差流规范如何影响所有可解释性工具
arXiv:2606.31963v1 Announce Type: new Abstract: Modern LLM workflows move coordinate-indexed objects across checkpoints: steering vectors, sparse auto…
揭秘大模型如何在超维叠加中完成纠错，读懂LLM计算的隐藏机制。
arXiv:2606.24964v1 Announce Type: new Abstract: Understanding the features of large language models (LLMs) is a central goal of interpretability. LLMs…
追溯大语言模型中可解释单元的训练数据源头，为理解模型行为提供全新归因方法
arXiv:2601.21996v2 Announce Type: replace-cross Abstract: While Mechanistic Interpretability has identified interpretable circuits in LLMs, their caus…
提出多Token归因方法，让长推理链LLM的可解释性更高效、更忠实，ICML Spotlight论文。
arXiv:2602.01914v2 Announce Type: replace Abstract: Token attribution methods provide intuitive explanations for language model outputs by identifying…
教皇Leo XIV通谕探讨AI时代的社会问题，类比工业革命，揭示LLM可解释性困境
Article URL: https://simonwillison.net/2026/May/25/encyclical-on-ai/ Comments URL: https://news.ycombinator.com/item?id=48275098 Points: 52 # Comments…