One of my favourite conferences (ALT) chaired by two of my favourite people (@thejonullman and Matus Telgarsky) at my favourite venue (@FieldsInstitute) in my favourite country!🇨🇦
Definitely submit and attend!!!
New award this year 👀
One of my favourite conferences (ALT) chaired by two of my favourite people (@thejonullman and Matus Telgarsky) at my favourite venue (@FieldsInstitute) in my favourite country!🇨🇦
Definitely submit and attend!!!
New award this year 👀 https://t.co/YzbXPq0cwl
I am excited to share our new monograph that dives deep into correlated noise mechanisms, one that enables real-world, production-grade differential privacy at scale, without compromising learning performance.
The project started with a discussion over lunch in Google's Canopy…
Textual steering vectors can improve visual understanding in multimodal LLMs!
You can extract steering vectors via any interpretability toolkit you like -- SAEs, MeanShift, Probes -- and apply them to image or text tokens (or both) of Multimodal LLMs.
And They Steer!
For this week’s NLP Seminar, we are thrilled to host @DeqingFu to talk about Closing the Modality Gap: Benchmarking and Improving Visual Understanding in Multimodal LLMs!
When: 5/22 Thurs 11am PT
Non-Stanford affiliates registration form (closed at 9am PT on the talk day):…
*Every single* cure for a disease ultimately flowed from basic exploratory research. Stopping basic research is like stopping the mountain rains and expecting rivers of cures to still flow. Examples:
1) studying saliva of Gila monster -> GLP1's
2) studying funghi -> first…
*Every single* cure for a disease ultimately flowed from basic exploratory research. Stopping basic research is like stopping the mountain rains and expecting rivers of cures to still flow. Examples:
1) studying saliva of Gila monster -> GLP1's
2) studying funghi -> first…
Our work (with @tianyi_zhou12@DeqingFu and Vatsal Sharan) published at NeurIPS 2024 already found that pretrained LMs do addition using modular arithmetic/“trigonometry” (we called these Fourier features). Indeed it is a clever mechanism.
Our work (with @tianyi_zhou12@DeqingFu and Vatsal Sharan) published at NeurIPS 2024 already found that pretrained LMs do addition using modular arithmetic/“trigonometry” (we called these Fourier features). Indeed it is a clever mechanism.
🚨 Introducing Hybrid Preference Optimization (HPO):
Aligning LLMs with human preferences just got more efficient! Our new approach combines online exploration with offline comparisons, achieving a provably optimal bound for hybrid RLHF. Read our paper: arxiv.org/pdf/2412.10616.
Super proud of @IITKanpur and Nitin Saxena for doing this.
Partha Halder supported me when I did Physics Olympiad. Nitin Saxena was one of the smartest people I have met / learnt from. This is an absolutely great turn of events.
timesofindia.indiatimes.com/education/news…
“mathematical breakthroughs are not powered solely (or even primarily) by ‘Eureka’ moments of genius, but are in fact largely a product of hard work” —Terry Tao
terrytao.wordpress.com/career-advice/…
796 Followers 399 FollowingAsst. Professor @Rutgers. Previously, privacy researcher @Apple, postdoc at @JohnsHopkins, @PennState, and a PhD student @UWaterloo
127 Followers 3K Following🧑💻 comparative programmer (Franted Backend Developer)Software engineer by day cashing in those bits and bytes 💸 Sharing all my l@ GitHub🌆
Developer
138 Followers 556 FollowingMasters in AI @USCViterbi '24|
Btech CS @IIITDelhi '22|
International Chess Player |
I can talk about Machine Learning all day😌
847 Followers 100 Followingtwt'ng and figuring out life :') | I rarely twt about tech here | SWE @Google | Former SWE Intern @Uber, @Google, https://t.co/voF1DCYaeP | CSE'23 @IIITDelhi
1K Followers 1K FollowingCS (HCC) PhD student @UMich | ex research @adobe | Human AI Interaction, Social Computing, Interaction Design | ruminating about des, tech & society🍂💫🌊
347 Followers 261 FollowingMachine Learning Scientist@eBay, Operations Research PhD@USC. Large-scale Optimization for ML models, Responsible AI (robust and fair ML), & Ranking Problems.
1K Followers 2K FollowingPhD student at @CornellCIS; ex-Research Fellow at @MSFTResearch India (@IndiaMSR); @IIITDelhi alum; All opinions are my own. he/him/his
47 Followers 39 Following4th year phd student @stanford, working on multigroup fairness notions, machine learning theory, and uncertainty quantification.
600 Followers 698 FollowingAssistant Professor at @WisconsinCS. Machine learning, privacy, and memorization. Postdoc @uwcse and PhD at Boston University.
1K Followers 123 FollowingMinker Professor of CS, ACM Fellow, Guggenheim Fellow,AAAS Fellow, Blavatnik Honoree, IEEE Fellow,EATCS Fellow working on Algorithms, Game Theory and BIG Graphs
2K Followers 76 FollowingGoogle Fellow, VP | Gemini Data Area Lead | Algorithms, GraphML, ML efficiency, Economics @ Google Research. Former MSR, Amazon, MIT PhD, Sharif Univ. BSc
8K Followers 411 FollowingDad and associate professor at @KAUST_News.
Formerly @BU_ece, @sbucompsc, @YahooResearch, @TTIC_Connect.
ML theory&practice, obsessed with history of science
50K Followers 884 FollowingAssistant professor (of mathematics) at the University of Toronto. Algebraic geometry, number theory, forever distracted and confused, etc. He/him.
3K Followers 433 FollowingAssistant Professor @JohnsHopkinsAMS, Optimization, PhD @Cornell_ORIE
Mostly here to share pretty maths/3D prints, sometimes sharing my research
11K Followers 723 Following"If there is not folly in the world, then the world itself is folly. You must understand that mistakes are not always regrets." - Paul Tobin, Bandette🤠
4K Followers 1K FollowingTheoretical CS prof@ U. Michigan-Ann Arbor. Opinions are mine & may evolve over time. r/t≠endorsement.
Soon moving to Bluesky: @mahdi.ch
he/him/his
6K Followers 314 FollowingUTCS is a recognized leader in creating the scientific knowledge and practical technologies exemplifying the digital revolution that defines the 21st century.
10K Followers 605 Followingprobability, statistics, metric spaces, Markov chains, freedom (social & academic), Israel, Jew stuff. opinions represent my employer & all other groups I'm in