🚨Final call: 2 AI Full Professorship openings at HPI / University of Potsdam (Berlin region in Germany) Happy to answer questions about what makes our institute really unique! Deadline: Jan. 15!! Application details: hpi.de/fileadmin/us... #MLSky

I really think where we went wrong was algorithmic feeds. by far the best and most fulfilling part of my BlueSky experience is the one list that includes friends (including ones I've never met IRL)

This seems fine

One of my favorite findings: Positional embeddings are just training wheels. They help convergence but hurt long-context generalization. We found that if you simply delete them after pretraining and recalibrate for <1% of the original budget, you unlock massive context windows. Smarter, not harder.

Please welcome Google's Open Source efforts to Blue Sky at @opensource.google!

Finally proceeded to deactivate+delete my Twitter account. I had kept it open but inactive as an archive of sort, frozen and unused, but given what X+Grok have become and now enable *by design*, it has become impossible to justify even that.

Grading and googling hallucinated citations, as one does nowadays, and now that LLMs have been around for a while, I've discovered new horrors: hallucinated journals are now appearing in Google Scholar with dozens of citations bc so many people are citing these fake things

The #ICLR2026 #OpenReview drama continues! 🙀 After the leak of anonymous reviewer identities, it is understandable that we need to mitigate potential effects this may have had. Still, it can feel disappointing for the majority of authors who played by the rules. Let's hope the ACs do a good job!

This is by a large margin the most serious problem and mistake in conference peer review I have seen in my career. Apparently a many people were aware of this and many could find out who their reviewers were. This probably has created a large number of unnecessary enmities. @iclr-conf.bsky.social

Summit on European Digital Sovereignty 🤔Any tangible results or is it all just talk? Sadly, politics runs slower than industry/academia. 👉🏽But discussions/declarations aren't meaningless. They set the conditions for taking sovereignty more seriously in the future.

Yann Lecun (@yann-lecun.bsky.social) has done a lot for the AI community during his years at Meta/Facebook. Let's hope that those efforts continue despite the departures of people like Yann and Soumith Chintala (@soumithchintala.bsky.social).

#AI@HPI Conference 2025 📌 Vom 2.–4. Dezember diskutieren Gesellschaft, Wirtschaft & Wissenschaft zusammen Strategien, um Europas Position im globalen KI-Wettbewerb zu stärken. Lest im Interview mit Gastgeber Prof. Gerald de Melo wie Forschung die #KI-Souveränität stärkt: hpi.de/artikel/wir-...

“When it comes to AI, European companies rarely play in the top league” – My perspective on why we need AI sovereignty. 👉🏽 hpi.de/en/article/w... 🚨Join us at the AI @ HPI Conference 2025 this December: hpi.de/en/ai-hpi-co...

www.envisionbox.org has been shortlisted for the Leo Waaijers Open Science price: ukb.nl/en/news/shor... @babajideowoyele.bsky.social @jamestrujillo.bsky.social @sarkadava.bsky.social @DavideAhmar @acwiek.bsky.social Amazing Markus Küpper made an animated video: www.youtube.com/watch?v=HduI...

Remembering Pushpak Bhattacharyya, known as Godfather of Indian NLP, former President of ACL + Director of IIT Patna. Despite his countless achievements, he remained humble and kind. He was one of the first people I met at an academic conference. May he rest in peace. www.cse.iitb.ac.in/memorial

Post-Doc positions at HPI! Feel free to reach out to me / check out nlproc.org/join/ for further general info. #mlsky 🤖🧠

you got it boss

OpenAI's Benedict Kerres at our @hpi.bsky.social Tech Leadership Conference: ChatGPT used for relationship and health advice a lot in Germany

Papers *written* and *reviewed* by AI 🤖 The Stanford Agents4Science conference is seeking research papers mainly authored by AI, which will then be peer-reviewed by multiple AI systems. Might seem odd to us now, but this might become common practice in the future. agents4science.stanford.edu

😮 Recent OpenAI / former DeepMind researcher talking about "mass migration as being worse for Europe than WW2 was" and extremist phantasies such as "large-scale remigration". x.com/RichardMCNgo...

In person (no streaming/zoom) sentence processing workshop at Potsdam with Tal Linzen, Brian Dillon, Titus von der Malsburg, Oezge Bakay, William Timkey, Pia Schoknecht, Michael Vrazitulis, and Johan Hennert: vasishth.github.io/sentproc-wor...

The University of Potsdam invites applications for 5 postdoc positions, incl. Cognitive Sciences, incl. NLP (esp. cognitive). These are fairly independent research positions that will allow the candidate to build their own profile. Dln June 2nd. Details: tinyurl.com/pd-potsdam-2... #NLProc #AI 🤖🧠

What is transformative about LLMs: Plurality (majority?) of students are no longer reading or writing papers Soon the majority of all known writing will be synthetic Videos of people now indistinguishable from the real thing Net new types of fraud with no known counter etc Take it seriously!

VEO3 Prompt: “a man doing stand up comedy in a small venue tells a joke (include the joke in the dialogue)” The video and audio are generated together.

Super excited Marin is finally out! Come see what we've been building! Code/platform for training fully reproducible models end-to-end, from data to evals. Plus a new high quality 8B base model. Percy did a good job explaining it on the other place. marin.community x.com/percyliang/s...

Wish I had a single @overleaf.com keyboard shortcut that would 1) append whatever I have in my clipboard to my bib file, 2) parse it, 3) extract the bibkey, and 4) paste it at my current cursor location in the editor.

Portugal is taking online disinformation and fake news seriously. As part of a broader attempt to encourage media literacy, it’s offering 15-18 year olds a free digital subscription to a range of respected newspapers and magazines as well as online upstarts that don’t spread lies. (Monocle)

Three year postdoc position in Cognitive Science, Potsdam, Germany: www.uni-potsdam.de/en/verwaltun...

Big news 7 years in the making www.iccl.ie/digital-data...

The Pope chose his name because he’s concerned about the effects of AI on human dignity, justice, and labor. www.theverge.com/news/664719/...

Scientists wanting to move to Europe from the US take note of this funding opportunity: erc.europa.eu/news-events/...

Come & work with me @hpi.bsky.social & @bmittelstadt.bsky.social & @cruss.bsky.social @oii.ox.ac.uk I am looking for 3 post docs on the governance of emergent tech. CS: tinyurl.com/yr5bvnn5 Ethics: tinyurl.com/yc2e2km4 Law: tinyurl.com/4rbhcndp Application deadline is 15.06.2025.

Microsoft is spending about 6x the annual budget of the entire NSF on building more server farms.

it’s here! a real Qwen3 model huggingface.co/Qwen/Qwen3-0...

Quick link to the paper: openreview.net/forum?id=aPT...

Had a great time presenting and talking to many of you at #ICLR2025.

I wrote something up for AI people who want to get into bluesky and either couldn't assemble an exciting feed or gave up doomscrolling when their Following feed switched to talking politics 24/7.

PhD Timeline xkcd.com/3081

Excited to be attending the @iclr-conf.bsky.social for the first time! I will be presenting AlignCLIP at the main conference and JinaCLIP-V2 from Jina AI at the SCI-FM workshop. Ping me if you want to talk about VLM/MLLMs at #ICLR2025.

Dear US-based misinfo/disinfo research community: I am looking to write about the impact of Friday night's NSF decision. I understand that this is a stressful moment, but also believe it is worth giving the public an understanding of what's at stake. docs.google.com/forms/d/1MQP...

One of my PhD students got their visa revoked. I know of other cases amongst my AI colleagues. This is not what investing in US leadership in AI looks like. www.aljazeera.com/news/2025/4/...

also got my first NSF grant terminated! what emoji do i put in my handle to signify this?

1/If EU has lost faith in US tech privacy for itself, how long until it loses faith in US tech for its societies… www.ft.com/content/20d0...

1. Now that China is cutting economic ties, new research on why China opted for technological self-reliance. The good stuff discussed below is down to my co-authors rather than me - but it _is_ good stuff. We have lots of NLP data that helps us map out China's debates about technological dependence.

🚨Llama 4 Is Out!🚨 2 out of 3 models just released - Scout: 109B / 17B active - Maverick: 400B / 17B active - Bohemoth: 2T / 288B active ai.meta.com/blog/llama-4...

So far, 2,135 people have responded to the poll Søren and I posted a few days ago. Of those, 94.4% replied “Yes” to being interested in officially presenting accepted @neuripsconf.bsky.social papers in Europe. (1/7)

Fantastic opportunity if you are interested in moving your lab to Montreal, Canada! Through this program, Chairholders will be awarded up to $1 million per year for a period of 8 years. Contact me if you are interested. www.nature.com/naturecareer...