Yi-han Sheu
@yihansheu
AI & Mental Health | Psychiatrist & Epidemiologist & ML | Instructor at Harvard & MGH | Opinions are my own. https://yihansheu.github.io/
Building a clinical prediction model involves more than just modeling. As a framework, I recommend always aligning the following four levels of abstraction: The clinical task – A clear definition of the clinical question, ensuring its relevance and significance. The estimand – A…
Nice!
Official results are in - Gemini achieved gold-medal level in the International Mathematical Olympiad! 🏆 An advanced version was able to solve 5 out of 6 problems. Incredible progress - huge congrats to @lmthang and the team! deepmind.google/discover/blog/…
We're excited to announce a second physical location for NeurIPS 2025, in Mexico City. By expanding our physical locations, we hope to address concerns around skyrocketing attendance and difficulties in obtaining travel visas that some attendees have experienced in the past few…
Looks promising
Energy-Based Transformers are Scalable Learners and Thinkers
With some abuse of analogy, Waymo is reminiscent of Deep Blue in chess. However, driving, unlike chess, is not a game with a fixed set of rules and finite states. The Waymo approach seems to impose a set of hard-wired constraints that prevent it from scaling. Of course, they can…
Apparently, LLMs today have capabilities and limitations that approximate human-level intelligence, although along dimensions different from those involved when a child develops cognitive functions into adulthood. That is a well-known, objective fact. What is the added value or…
The Illusion of Thinking in LLMs Apple researchers discuss the strengths and limitations of reasoning models. Apparently, reasoning models "collapse" beyond certain task complexities. Lots of important insights on this one. (bookmark it!) Here are my notes:
Today is the start of a new era of natively multimodal AI innovation. Today, we’re introducing the first Llama 4 models: Llama 4 Scout and Llama 4 Maverick — our most advanced models yet and the best in their class for multimodality. Llama 4 Scout • 17B-active-parameter model…
Nice
New Anthropic research: Tracing the thoughts of a large language model. We built a "microscope" to inspect what happens inside AI models and use it to understand Claude’s (often complex and surprising) internal mechanisms.
About a month after Donald Trump took office, almost all grant-review meetings remain suspended at the US National Institutes of Health, preventing the world’s largest public funder of biomedical research from spending much of its US$47 billion annual budget.…
A couple reflections on the quantum computing breakthrough we just announced... Most of us grew up learning there are three main types of matter that matter: solid, liquid, and gas. Today, that changed. After a nearly 20 year pursuit, we’ve created an entirely new state of…
🚀 Introducing NSA: A Hardware-Aligned and Natively Trainable Sparse Attention mechanism for ultra-fast long-context training & inference! Core components of NSA: • Dynamic hierarchical sparse strategy • Coarse-grained token compression • Fine-grained token selection 💡 With…