I am leaving X in favor of BlueSky. Since Elon took over, X departed from Twitter's values and turned into a dystopian platform. I have three two main reasons:
(1) Zero content moderation is unacceptable.
(2) Right-wing content skyrocketed.
Details in the 🧵⬇️
(2) Right-wing content skyrocketed. This was strategic due to the 2024 presidential election. All users were at the mercy of Elon's personal interests. If that is
not the definition of dystopia, what is? There is ample evidence for this: https://t.co/XsG8xmKAJF
I am leaving X in favor of BlueSky. Since Elon took over, X departed from Twitter's values and turned into a dystopian platform. I have three two main reasons:
(1) Zero content moderation is unacceptable.
(2) Right-wing content skyrocketed.
Details in the 🧵⬇️
(1) Social media platforms have responsibilities akin to
journalistic formats. They dont *generate* fake news, but they should be accountable if they flush it to my newsfeed. If the NYT had a fake news headline, the publisher would resign. Listen to Harari https://t.co/srEvDm5pt5
📖Book Chapter Alert! “Language Models in Molecular Discovery” (https://t.co/CfuvBRpnuq) is a deep dive into how language models can be used to accelerate molecular discovery, highlighting their strengths and weaknesses.
Our ICML paper: ML + quantum computing + optimal transport = 🚀!
Learn distributions conditionally with quantum. Including a 24-qubit hardware run on the assignment problem (predicting doubly-stochastic matrices) @IBMResearch
📖 https://t.co/oSKJxT8Iew
📹 https://t.co/tAYFReAWzv
Thrilled to share that our paper 'Impact of early visual experience on later usage of color cues', combining experimental data from children treated for congenital blindness and results of computational simulations, is now out in @ScienceMagazine! https://t.co/D78BCzqRQw 🧵
Consider these fantastic opportunities in #Lausanne if you're into ML for Biomedicine and are looking for a PhD, PostDoc or master thesis!
I can speak from experience and it has been super inspiring to work with @marianna_raps!
📢 We are hiring! 📢
Are you a MSc/PhD👩🎓👨🎓 passionate about #AI#ML in #cancer, looking for your next career step?
Join our newly-founded AI/ML for Biomedicine group part of the Biomedical Data Science Center at @unil and @CHUVLausanne in beautiful Lausanne!🇨🇭🏔🌈
👇
In our new paper in @Nature_NPJ we propose to use **z-scored** drug response measures (IC50, AUC etc). We show that conventional metrics hamper the development of personalized prediction models 🤯
Thanks to all collaborators!
New in #npjPrecisionOncology: Drug response models overlook nuances of drug effects between cancer subtypes. The solution? Use z-scored measures to unlock more personalized drug response predictions @jannisborn @MariannKruithof @Katja_HB4@PChouvardas
https://t.co/OZBsheyZVG
Interested in Uncertainty Quantification for Sequence Prediction?
Come check out our poster at @RealAAAI today. We add rigorous uncertainty quantification to Beam Search using Conformal Predictions!
@nickgermann@IBMResearch
Introducing Clipboard-to-SMILES Converter: a macOS app for effortless conversion of screenshots into molecular structures (SMILES, SELFIES, etc.) right from your clipboard, complete with a convenient history feature!
Check out our paper and download it: https://t.co/gYmDT79ami
We have now completed our analysis of new materials reported in the Google Deepmind / Berkeley autonomous lab paper. My own initial analysis is in the quote tweet.
Happy to have worked with @SchoopLab to jointly put together a comprehensive analysis, now available on @ChemRxiv.
This thread is my personal view after having looked at the detail of each of the materials reported 🧵
https://t.co/7j23VCUVTK
Entering 2024 as a Research Scientist at @IBMResearch Europe💃 thrilled to join a team of pure🔥 and to finally be a full member of the crew. Looking forward to the great things coming ahead 🤓
Our paper got accepted at #AAAI24! 🎉
We propose two new
sequence generation algorithms with "error bars", by adapting beam search to conformal predictions.
If you use LLMs for science, like predicting molecules or proteins that verify some conditions, check it out!
4/4 “Characterizing pre-trained and task-adapted molecular representations” — #UniReps workshop (Friday)
TL;DR: A post-hoc quality evaluation method for representation learning models and domain adaptation methods that is task and modality-agnostic
We're in 2nd half of #NeurIPS2023, but still much exciting research ahead!Thrilled to present some work from #AI4SD team @IBMResearch. Topics cover quantum computing, foundation models, optimal transport, digitization of lab workflows, language models in molecular design & more⬇️
3/4 “Language Models in Molecular Discovery” — AI4Science workshop (Saturday, brought to you by @niklexical)
TL;DR: How to leverage language models to accelerate molecular discovery, despite their limitations.