Introduction to Neural Networks and Seq2Seq Learning

During the summer semester of 2026, I took Introduction to Neural Networks and Sequence-to-Sequence Learning course. As you see, the main content is from perceptrons to modern sequence models. I started learning machine learning through Li Mu’s courses. I spent several weeks systemically studying his materials and organized the corresponding code in the repo. Also, I had already used neural networks in multiple projects before taking this course. I had trained language models, vision models, vision-language models, reinforcement-learning agents and even vision-language-action models… ...

July 15, 2026 | 3419 words | Author: Tan Ke

Thinking Like Transformers

Paper-reading notes: RASP
December 7, 2025 | 273 words | Author: Tan Ke

It’s All Connected: A Journey Through Test-Time Memorization, Attentional Bias, Retention, and Online Optimization

Paper-reading notes: MIRAS
December 6, 2025 | 923 words | Author: Tan Ke

On the Representational Capacity of Neural Language Models with Chain-of-Thought Reasoning

Paper-reading notes: On the Representational Capacity of Neural Language Models with Chain-of-Thought Reasoning
December 1, 2025 | 462 words | Author: Tan Ke

What Formal Languages Can Transformers Express? A Survey

Paper-reading notes: What Formal Languages Can Transformers Express? A Survey
November 30, 2025 | 327 words | Author: Tan Ke

ATLAS: Learning to Optimally Memorize the Context at Test Time

Paper-reading notes: ATLAS
November 29, 2025 | 628 words | Author: Tan Ke

Solving olympiad geometry without human demonstrations

Paper-reading notes: AlphaGeometry
November 28, 2025 | 522 words | Author: Tan Ke

Formal Mathematical Reasoning A New Frontier in AI

Paper-reading notes: Formal Mathematical Reasoning A New Frontier in AI
November 27, 2025 | 347 words | Author: Tan Ke

Titans: Learning to Memorize at Test Time

Paper-reading notes: Titans
November 26, 2025 | 916 words | Author: Tan Ke

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Paper-reading notes: DeepSeek-R1 - Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
November 4, 2025 | 2299 words | Author: Tan Ke