Press ESC to close Press ⌘K or Ctrl+K to open

用行动赢得尊重

Alan Hou

Notes on coding, AI, and technology.

记录编程、人工智能与技术探索。

Recent Posts / 近期文章

View all posts

Counterfactual World Models: Keep the Disturbance, Change the Action

换个动作,留住同一阵风

Mapping the AI Supply Chain: Relationships Reveal Bottlenecks

看见关系,才看见瓶颈

A User Simulator Must Pass More Than a Turing Test

像用户,不等于懂用户

SemPhys: Semantic-Physical Co-Learning for Humanoid Motion Generation via Differentiable World Models

SemPhys:基于可微世界模型的具身人形机器人语义-物理协同生成

Test-Time Symmetrization: Scaling Laws for Model-Agnostic Learning under Structure

测试期对称化:结构对称性下的模型无关测试期算力缩放法则

Check Token: Real-Time Self-Verification and Precise Truncation in LLM Reasoning

校验 Token:大语言模型推理流的实时自检与精准截断重写

Policy Optimization With Multiple Verifiers via Pareto Rankings

多重验证器下的策略优化:基于帕累托偏序排名的无标量化强化学习

What to Practice Next? Latent Profile-Guided Task Selection for Skill-Evolving LLM Agents

接下来练什么?基于潜在能力图谱的自演化智能体任务选择

Recovered or Memorized? Investigating Dynamical Consistency and Faithfulness in World Model Latents

真学到了物理规律还是单纯死记硬背?世界模型潜空间的动力学一致性与保真度检验

Series / 系列文章