David Rügamer

After yesterday's Salon de Refusés paper presentation and today's EIML workshop paper presentation with @jkobialka.bsky.social, we will conclude with an oral by @erpel.bsky.social at the MedEurIPS workshop tomorrow 🔥 Very happy that we had so many great opportunities to present our research! ❤️EurIPS

EurIPS was pretty fun! Thanks to @lacerbi.bsky.social , @conorhassan.bsky.social , and all organizers of the Amortized ProbML team for the great workshop and for inviting me as a speaker! 🙌 Also great to be able to present our spotlight paper on sparsity on the main stage 😍

If you are at @euripsconf.bsky.social and you want to work with me on probabilistic ML then get in touch!

What coding with an LLM feels like sometimes.

#MCML researchers have contributed more than 40 papers to NeurIPS 2025! Check them out: mcml.ai/news/2025-11...

Congrats to our researchers! 🎉 #NeurIPS2025

bsky.app/profile/davi...

OpenReview's Statement Regarding API Security Incident - November 27, 2025

They are already targeting the "40-questions 40-weaknesses reviewer" on Xiaohongshu, and he has since raised his score.

which seems a bit naive.... the dataset was even on Hugging Face in the meantime

Here is the official @iclr-conf.bsky.social statement

OpenReview had a bug that allowed crawling reviewer names for ICLR (and apparently for all previous and ongoing conferences), and it had been known since Nov 12 ................

Paper Copilot is also working now papercopilot.com/statistics/i...

Reviewers are allowed to use LLMs but need to report it in their review. Not sure if this is visible to authors but its not super informative anyway (what I see as an AC is an "LLM usage: yes" entry for some of them)

📣 #ICML tutorials: We want to know what *you* would like to learn. This year, Adam White and I are calling for nominations of topics and/or presenters.

Until December 7th, you can send us your suggestions, and we will use them to shape the program.

icml.cc/Conferences/...

Cross-posting from X 👇 (cc @iclr-conf.bsky.social — don’t forget about the butterflies!)

The next generation of startup ecosystems: a benchmarking This report is the conclusion of a data-driven analysis of 201 tech ecosystems. This set of actionable benchmarks is meant as a tool to help ecosystems understand and measure their maturity and…

Munich is now the #2 AI ecosystem in Europe (after Paris) and #11 globally, per Dealroom’s Global Tech Ecosystem Index 2025! 🇩🇪🤖🇫🇷
Proud of the growing strength of the Bavarian AI ecosystem 👉 baiosphere.org
�� dealroom.co/tech-ecosystem-index-2025�5
#AI #Munich #baiosphere

The blog post is available: blog.arxiv.org/2025/10/31/a...

I have not announced it widely but I will be starting a professorship in TU Darmstadt (Germany) next year. I will start accepting PhD students from Sep 2026 (but I will keep my group and position in Tokyo as well).

Here is a first call for PhD positions at our new Center of Excellence at TUDa […]

𝗖𝗮𝗹𝗹 𝗳𝗼𝗿 𝗳𝘂𝗹𝗹𝘆 𝗳𝘂𝗻𝗱𝗲𝗱 𝗣𝗵𝗗 𝗣𝗼𝘀𝗶𝘁𝗶𝗼𝗻𝘀: We are offering several PhD positions across our various research areas, open to highly qualified candidates.
‼️ The application portal will be open from 15 October to 14 November 2025.

Find out more: mcml.ai/opportunitie...

Works like a charm for structured sparsity too 🔥 Tested on a variety of architectures: sparse attention heads, sparse inputs, sparse conv filters, sparse NODEs,... And no setup changes needed, just a regularized gating variable! Spotlight @NeurIPS + Oral @EurIPS 🤩
Preprint: arxiv.org/abs/2509.23898

Any updates on the Salon des Refusés? Do you also accept position papers?

🚀 RTG-METEOR is hiring!

Apply now for Doctoral (4y) or Postdoc (3y) positions (TV-L E13, 100%) at LMU & TUM.
Research focus: bridging Machine Learning & Control Theory for complex dynamical systems.

📅 Deadline: Nov 13, 2025
📍 Start: Apr 1, 2026

www.portal.graduatecenter.lmu.de/ocgc/meteor

The Journal-to-Conference track is back at AISTATS🌴!

The track invites authors of papers published or accepted after January 1st, 2024 in AoS, Biometrika, JASA, JMLR and JRSS B to present their work as posters at #AISTATS2026 in Tangier, Morocco (May 2nd-5th 2026).

📅 Deadline: 31 January 2026

I am not sure if they are opposed to this, but "[t]he Mexico City location will host approximately five hundred attendees". Definitely seems too small, given that China alone had almost 1,600 papers last year: papercopilot.com/paper-list/n... (click on Aff. Country)

@zeynepakata.bsky.social Maybe suggest that they accept EurIPS Copenhagen as an official third location. I think many PhD students would be happy to "just" present in Europe :)

Right. Maybe ICLR should be more precise in their statement then (e.g., it currently only refers to 'hidden' injections).

Or would I still be accused of collusion, since I am technically influencing the review---even though what I am really doing is making public that I just want to be treated fairly?

I was wondering: What if the prompt is not hidden and I simply want to protect myself from lazy reviewers? For instance, I could openly include a protection sentence in the appendix that instructs an LLM to add a subtle identifier, allowing authors and chairs to detect LLM-generated reviews.

@jwenger.bsky.social : you can find it under 'Nomination' here virtual.aistats.org/Conferences/...

Symmetry and Geometry in Neural Representations on X: "‼️‼️ DEADLINE EXTENDED ‼️‼️ We are extending the NeurReps submission deadline to next Friday, August 29, 11:59 pm AOE. Follow these instructions to get your papers in: https://t.co/A3jYmIlTVH Best paper in each category will receive a $500 cash prize 🏆" / X ‼️‼️ DEADLINE EXTENDED ‼️‼️ We are extending the NeurReps submission deadline to next Friday, August 29, 11:59 pm AOE. Follow these instructions to get your papers in: https://t.co/A3jYmIlTVH Best paper in each category will receive a $500 cash prize 🏆

Some tweet-to-flap mapping: NeuReps (www.neurreps.org) deadline extended by a week: x.com/neur_reps/st...