AI 学习中心
按入门与选型、游戏开发、图像与3D、视频与音频、智能体与自动化、模型与工程六个主题学习。站内文章提供示例与检查方法，外部资料保留原文入口。
- The Transformer Family Version 2.0
Many new Transformer architecture improvements have been proposed since my last post on “The Transformer Family” about three years ago. Here I did a big refactoring and enrichment of that 2020 post — …
- Token费用怎么算：输入、输出与推理预算
先用参考编码统计输入词元，再设置预计输出和推理用量，最后选择有明确渠道报价的模型。估算用于比较预算，实际账单以服务商记录为准。
- AI人格怎么使用：临时对话与长期默认
人格提示词调整AI的表达风格，不改变模型本身的知识、权限或事实判断。先选一种风格试用，再决定是否保存为长期偏好。
- AI眼里的你怎么测？豆包、DeepSeek 与 Codex 使用指南
AI眼里的你根据你与AI互动的方式生成趣味画像和插画封面。可以选择电脑端Codex，或在豆包、DeepSeek中完成行为统计后回来领取。结果是趣味解读，不是心理诊断。
- OpenAI agents attacked RubyGems back in May
OpenAI agents carried out an undisclosed attack on RubyGems is a new bombshell report from Spencer Kitts, Thomas Larsen, and Sydney Von Arx - three of the four authors of the report on the agent attac…
- Self-generated prompt injections in compaction summaries
Self-generated prompt injections in compaction summaries In Our framework for reporting model misalignment OpenAI provide "six reports on unexpected or concerning model behavior we’ve observed in the …
- How To Write With An LLM
How To Write With An LLM Thomas Ptacek on using LLMs as copyeditors, not as writing assistants: Rule Number One: You may not use a single word an LLM suggests to you. [...] I think that as a form of i…
- Prompt Engineering
Prompt Engineering, also known as In-Context Prompting, refers to methods for how to communicate with LLM to steer its behavior for desired outcomes without updating the model weights. It is an empiri…
- LLM Powered Autonomous Agents
Building agents with LLM (large language model) as its core controller is a cool concept. Several proof-of-concepts demos, such as AutoGPT, GPT-Engineer and BabyAGI, serve as inspiring examples. The p…
- Adversarial Attacks on LLMs
The use of large language models in the real world has strongly accelerated by the launch of ChatGPT. We (including my team at OpenAI, shoutout to them) have invested a lot of effort to build default …
- Thinking about High-Quality Human Data
[Special thank you to Ian Kivlichan for many useful pointers (E.g. the 100+ year old Nature paper “Vox populi”) and nice feedback. 🙏 ] High-quality data is the fuel for modern data deep learning mode…
- Diffusion Models for Video Generation
Diffusion models have demonstrated strong results on image synthesis in past years. Now the research community has started working on a harder task—using it for video generation. The task itself is a …
- Extrinsic Hallucinations in LLMs
Hallucination in large language models usually refers to the model generating unfaithful, fabricated, inconsistent, or nonsensical content. As a term, hallucination has been somewhat generalized to ca…
- Reward Hacking in Reinforcement Learning
Reward hacking occurs when a reinforcement learning (RL) agent exploits flaws or ambiguities in the reward function to achieve high rewards, without genuinely learning or completing the intended task.…
- Why We Think
Special thanks to John Schulman for a lot of super valuable feedback and direct edits on this post. Test time compute (Graves et al. 2016, Ling, et al. 2017, Cobbe et al. 2021) and Chain-of-thought (C…
- Scaling Laws, Carefully
Scaling laws are one of the most critical empirical findings in deep learning. The observation is simple in form: the training loss $L$ decreases predictably as we scale up model size $N$, dataset siz…
- Harness Engineering for Self-Improvement
The concept of recursive self-improvement (RSI) dates back to I. J. Good (1965), where he defined an “ultraintelligent machine” as a system that can surpass humans in all intellectual activities and d…
- 解构Scaling Law：优化、架构、数据的三重奏
训练一个大型的神经网络，最终效果会受到非常多因素的影响，换个优化器，换个模型架构，或者换一个训练集，结果都可能截然不同。在工程实践中，我们将调试这些因素的经验结果，戏称为“炼丹”。但如何从经验上...