iiinova — where your digital self keeps evolving

Reposted by William Jurayj

Kate Silbaugh, Professor at BU Law @katesilbaugh.bsky.social · 22/02/2026

'Corporate Childrearing' is forthcoming in Duke Law Journal (26-27). Corporations play a huge role in children's identity formation. The piece reshapes the family law triangle into a square to make their influence and their intrusion on family relationships explicit. papers.ssrn.com/sol3/papers....

152

Reposted by William Jurayj

JHU Computer Science @jhucompsci.bsky.social · 02/07/2025

JHU computer scientists including @williamjurayj.bsky.social propose a method that allows #AI models to spend more time thinking through problems & uses a confidence score to determine when the AI should say "I don't know" rather than risking a wrong answer, which is crucial for high-stakes domains.

hub.jhu.edu

Teaching AI to admit uncertainty

Johns Hopkins researchers show how different "odds" can teach AI models to admit when they're not confident enough in an answer

011

Reposted by William Jurayj

Mustafa Suleyman @mustafasuleymanai.bsky.social · 19/03/2025

You can't just be right, you have to know you're right. Good advice for LLMs, according to new Johns Hopkins research. Sometimes no answer is better than a wrong one - life or death choices in medicine, for example, or big financial decisions. 🧵

a 3D graph with the X axis of compute budget, Y axis of accuracy, and Z axis of confidence threshold. The chart shows that accuracy increases with higher compute and confidence thresholds, though the trade-off tends to be fewer questions answered overall.

1125

William Jurayj @williamjurayj.bsky.social · 20/02/2025

🚨 You are only evaluating a slice of your test-time scaling model's performance! 🚨 📈 We consider how models’ confidence in their answers changes as test-time compute increases. Reasoning longer helps models answer more confidently! 📝: arxiv.org/abs/2502.13962

11710

William Jurayj @williamjurayj.bsky.social · 06/12/2024

I’d say a key factor is whether a person’s put in a good faith effort to be right for the right reasons. But I’m to other explanations!

050

Reposted by William Jurayj

Marc Marone @marcmarone.com · 23/11/2024

I noticed a lot of starter packs skewed towards faculty/industry, so I made one of just NLP & ML students: go.bsky.app/vju2ux Students do different research, go on the job market, and recruit other students. Ping me and I'll add you!

10117654