Yi-han Sheu
@yihansheu
Followers
72
Following
359
Media
8
Statuses
118
AI & Mental Health | Psychiatrist & Epidemiologist & ML | Instructor at Harvard & MGH | Opinions are my own. https://t.co/CDvC1MnNhv
Joined September 2013
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
0
0
3
“Everything should be made as simple as possible, but not simpler”.
0
0
1
AI can perform many tasks—either autonomously or in an assistive role—but it remains most valuable to those with a strong foundation of knowledge and a willingness to learn. When these tools are still imperfect, that foundation allows us to assess the credibility of their.
0
0
2
GPT-5 feels like 4o+o4 with an auto KVM switch and just a little bit more.
0
0
0
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!
0
0
1
RT @NeurIPSConf: We're excited to announce a second physical location for NeurIPS 2025, in Mexico City. By expanding our physical locations….
0
81
0
Looks promising.
0
0
2
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.
0
0
1
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:
0
0
0
RT @AIatMeta: Today is the start of a new era of natively multimodal AI innovation. Today, we’re introducing the first Llama 4 models: Lla….
0
2K
0
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.
0
0
0
Our new paper on continuous-time, dynamic suicide attempt risk prediction is now published online at @npjDigitalMed. Using a large-scale electronic health records database, we adapt neural ODEs—naturally suited for modeling evolving latent states like suicidal tendencies and
3
5
30
RT @Nature: About a month after Donald Trump took office, almost all grant-review meetings remain suspended at the US National Institutes o….
0
424
0
RT @satyanadella: A couple reflections on the quantum computing breakthrough we just announced. Most of us grew up learning there are th….
0
19K
0
RT @deepseek_ai: 🚀 Introducing NSA: A Hardware-Aligned and Natively Trainable Sparse Attention mechanism for ultra-fast long-context traini….
0
2K
0