Sneak peak!

Just recorded a demo
youtu.be/3sKIoVp8QRw

Some alpha users found a ton of bugs, so beta release is a bit delayed till next week at least.

You can still sign up @ dataline.app!

It's already open source but not ready for full public release yet 😁

Elena Glassman Hussein Mozannar Minsuk Chang This is such a great thread!! Two classic readings to understand what “theory” means for HCI are: link.springer.com/book/10.1007/9… and dl.acm.org/doi/10.1145/34…

Beyond the quality of our decisions, how will AI assistance affect us -- our growth and improvement, enjoyment, collaboration, or our agency in the workplace? The current design of AI assistance does not consider human-centric objectives; we need methods to account for them.

Super cool from our lab for selectively decoding between multiple LLMs at each token!

Our paper on adapting Copilot from human feedback is at #AAAI24 !
One of the secret sauces to making GitHub Copilot work is knowing when to show suggestions, our paper presents a motivated approach (CDHF) to achieve this using feedback data
arxiv.org/pdf/2306.04930…

We show in our work that LLM augmentations can improve clinical note readability! I was mostly involved with quantitative evaluation, but there are fascinating insights about LLM errors and qualitative insights in interviews with breast cancer survivors arxiv.org/abs/2401.09637

👩‍⚕️💻 Our physician-machine partnerships paper is out in Nature Medicine today!

Experiment w/ 1000+ dermatologists & PCPs shows how AI assistance can boost diagnostic accuracy but reveals limits to addressing physician bias w/ fair classifiers

nature.com/articles/s4159…

Thread 👇

A new onboarding technique can help workers collaborate more effectively with AI assistants. The process finds situations where the human trusts the AI either too much or too little, then develops rules about them and creates training exercises. mitsha.re/cKP350Qqw2j

Start your new year the right way by participating in our web-based coding study!
The study involves solving Python tasks with AI for 35mins, 15$ Amazon gift card compensation, register interest with link: forms.gle/Pj6Ahy5Wf7JYMN…

Added rigor to machine learning experiments: pre-register your training/testing procedure (how you will split data, seed, features, ...)

I've been thinking about this for a while, happy to see a paper discuss this formally!

Our NeurIPS '23 paper by Hussein Mozannar and colleagues at MIT-IBM Watson AI Lab featured on MIT homepage today. Article here --
news.mit.edu/2023/automated…

New research at NeurIPS (will be there next week) on front page of MIT news.mit.edu/2023/automated…

I'm recruiting PhD students for my lab at Johns Hopkins!

Please apply if you're interested in reliable ML / causal inference for decision-making in healthcare. See my website (moberst.com) for more info.

Deadline 12/15. Retweets welcome :)
cs.jhu.edu/academic-progr…

🤖 Join Jason Fries Jason Alan Fries, Parisa Rashidi Parisa Rashidi, Hussein Mozannar Hussein Mozannar, & Rahul Thapta at #ML4H2023 for a roundtable on Health AI Collaborations, Deployment, and Regulation. How can we forge effective clinician-AI collaborations and shape AI policy?

I just published a book The MIT Press @mitpress.bsky.social that looks at 10 yrs of innovative work MIT CSAIL & the future of CS. Big tx to @MIT prof/coauthor Daniela Rus & so many others for making it happen!

More info: bit.ly/4a07ew3
Read: bit.ly/3Gl6V1q
Buy: bit.ly/3uyBFJM

Great news!! The deadline for the LebNet's Tech Fellows Program has been extended to November 30! Hurry up and apply forms.gle/yFqDnfcQit3BFa…

Finding 3: A better understanding of Y|X-shifts gives better data-centric interventions.
- A tool to identify risky regions (share similar ideas with recent work by Hussein Mozannar)
- Inspire better interventions: Efficient Data Collection & Feature Collection

Check this: