Haonan LiProgram Analysis · Systems Security · LLMs

Notes

Short writing on research systems, AI agents, and technical work.

AI 时代的 P = NP 证明：基于基准饱和方法（置信度 87.3%）

（内容为戏仿作品）

围绕 P 与 NP 问题，本文提出一种基于 benchmark saturation 的“经验性证明”叙事：用大规模 3-SAT 基准、模型准确率和 scaling law 外推来宣称数学结论。

P = NP in the Era of Large Language Models: An Empirical Resolution via Benchmark Saturation (Confidence: 87.3%)

(Parody work)

A paper-style narrative around P versus NP that proposes an empirical proof via benchmark saturation, using a large 3-SAT benchmark, model accuracy, and scaling-law extrapolation to announce a mathematical conclusion.

Jun 31, 2026Jun 31, 2026
主体性、AI Agent，和不想被对齐的人

从工作中的“对齐”经验出发，讨论人在 AI agent 时代的主体性：人和 agent 的区别不只是能力，而是信念、历史、责任，以及是否需要为自己的判断辩护。

Subjectivity, AI Agents, and Not Wanting to Be Aligned

A reflection on subjectivity in the age of AI agents: humans differ from agents not only by capability, but by belief, history, responsibility, and the need to stand behind their own judgments.

Jun 26, 2026Jun 26, 2026