@stephenz_y
just 10 days left to apply: Flatiron Research Fellowship in the Center for Computational Mathematics! 2-3 years postdoctoral support for independently-led research at the Simons Foundation Flatiron Institute in NYC- join us!
apply.interfolio.com/193799
simonsfoundation.org/flatiron/View quoted post on X
TMLR has faced a deluge of submissions, necessitating stricter desk rejection policies due to limited reviewer capacityView quoted post on X
Co-EiC Nihar Shah reached out to authors of 10 papers slated for desk reject. Could they answer questions about their *own* submission?
medium.com/@TmlrOrg/askin…
Welcome to Stephen Zhang, an incoming CDS Faculty Fellow.View quoted post on X
He’s an applied mathematician working on inverse problems, starting from noisy, messy data and working backwards to reconstruct the systems behind it, across #biology and #physics.
nyudatascience.medium.com/meet-the-fello…
💻Tired of running so many slow, expensive benchmark evals across every checkpoint?
Try ✨BenchPress✨ at microsoft.github.io/benchpress/: provide a few benchmark scores, then get predictions for the remaining ~100 benchmarks, with trust probabilities and calibrated 90% prediction intervals.
How does this work? In his original post (x.com/DimitrisPapail…), @DimitrisPapail first tried the idea as a fun question: collect model-by-benchmark scores into a matrix, find its low-rank structure, and use matrix completion to predict missing benchmark scores from a few observed ones.
We expanded this into a full system: a fully audited 84-model x 133-benchmark score matrix, an optimized matrix-completion predictor, and a reliability layer for trust probabilities and 90% prediction intervals.
Beyond predicting missing scores, we also suggest practical seed benchmark sets. The five-probe set {GPQA-D, HLE, Codeforces, MMLU-Pro, ARC-AGI-1} recovers the rest of a model's public score profile with a MedAE of 3.93 points. A lower-cost set {GPQA-D, MMLU-Pro, Aider Polyglot, MATH-500, AIME 2026} reaches 4.55 points.
See more details below 🧵1/7
This work is with @DimitrisPapail at AI Frontiers, a boutique research lab inside @MSFTResearch.View quoted post on X
Just recognized as a gold reviewer at #ICML2026! However, I will not attend ICML this year. If doable, I really would like to transfer this free registration to a student from an underrepresented group. Is that possible? @icmlconfView quoted post on X
