🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Excited to share that our paper got accepted at #ICML2025!! 🎉 We challenge Vision-Language Models like OpenAI’s o1 with Bongard problems, classic visual reasoning challenges and uncover surprising shortcomings. Check out the paper: arxiv.org/abs/2410.19546 & read more below 👇

Weil es wichtig ist, wie das zukünftige Bundesdigitalministerium gestricktsein wird, einfach einmal reinhören: The Gesamtschau Podcast open.spotify.com/show/4psTvnCyo…

An interesting paper by Eran Rosenbluth and Martin Grohe, arxiv.org/abs/2505.00291, shows a connection between polynomial-time algorithms and recurrent GNNs.

Horizontal, vertical, hybrid? Data heterogeneity is tough in federated learning! 🔥In "Scaling Probabilistic Circuits via Data Partitioning" - accepted UAI 2025 - we introduce a unified view through aggregation of learned client distributions: arxiv.org/abs/2503.08141

Ich freue mich, in der aktuellen Beyond the Obvious-Kolumne von Dr. Daniel Stelter im Handelsblatt zum neuen Digitalministerium aufzutauchen. Digitalisierung braucht endlich Rückenwind – jetzt kommt es auf kluge Umsetzung und echten Gestaltungswillen an. 👉 buff.ly/Opf5HdQ

Wer tiefer einsteigen will: Im Podcast „Eine letzte Chance“ spricht Dr. Daniel Stelter über das Digitalministerium – mit Bezug zu unserem Whitepaper zur digitalen Staatlichkeit. Hören lohnt sich! 🎧 👉 buff.ly/orPZ7tj

Would love to know what fraction of 25K+ #NeurIPS2025 submissions are resubmissions. Why can't we provide a "way out" for borderline papers so that folks can move on if they want to? The amount of time that we as a community spend on reviewing papers to then reject them is absurd

You may know batch-norm, but did you know it is a lightweight catalyst for open world learning? In "BOWL: A Deceptively Simple Open World Learner" we leverage BN stats to enable OOD detection & rapid active + continual learning! arxiv.org/abs/2402.04814 - accepted CoLLAs 2025🔥

aaai.org/aaai-launches-…

Submit to our #ICML2025 workshop on “The Impact of Memorization on Trustworthy Foundation Models” icml2025memfm.github.io The deadline is on May 27th. Understanding unintended memorization is essential to building trustworthy foundation models. Join us in Vancouver on July 19th

Our ICML & RLC workshops welcome contributions using programmatic representations as policies, reward functions, skill libraries, task generators, environment models, etc., to improve interpretability, generalization, efficiency, & safety in agent learning & RL! Please retweet 🙏

Online AI detector says Declaration of Independence is 99.99% written by AI. Which chatbot did Founding Fathers use?

"Reasonable AI" got selected as a cluster of excellence! hessian.ai/reasonable-and… Overwhelmingly happy to be part of RAI + work with the smart minds at TU Darmstadt & hessian.AI & see my new home at Uni Bremen achieve a historic success in the excellence strategy!

Den Willen für eine #BrainGain-Initiative sehe ich nach Trumps Harvard-Eklat in Deutschland bisher nicht: 539 Mio Euro gibt’s für die neue Exzellenzstrategie – verteilt auf 43 Unis. Aber für die Senkung der Gastro-Steuer sollen jährlich 4 Mrd fließen. Nicht nachvollziehbar!👇

🚀 TU Darmstadt leads the new Cluster of Excellence "Reasonable AI" – advancing trustworthy, efficient & adaptive AI grounded in common sense. A big thank you to the entire team, the university, and the state of Hesse for their tremendous work & support! buff.ly/N7xLKfW

New paper Transactions on Machine Learning Research: Felix Divo compares properties & performance of 24 forecasting models on financial data from ACATIS Investment. Deep learning models, as expected, perform well. Cherry on top: model expectations align with a human analyst tinyurl.com/4vp5y7t8 Kristian Kersting

🚀 Do you want to nudge your Bayesian Optimization into the right direction effectively? Then check out our new work Hyperparameter Optimization via Interacting with Probabilistic Circuits which got accepted at this year's AutoML_conf!

🙏 for the invitation to give a talk at the Causal Neuro-Symbolic AI (Causal-NeSy) Workshop at ESWC. Integrating causality, neuro-symbolic reasoning, and machine learning to build more robust, interpretable, and generalizable AI systems is so important.

🚀 New Preprint We introduce JQL: a highly efficient, modular pipeline for multilingual pre-training data curation. 📄 𝐀𝐫𝐗𝐢𝐯: arxiv.org/abs/2505.22232 🤗 𝐇𝐮𝐠𝐠𝐢𝐧𝐠 𝐅𝐚𝐜𝐞: huggingface.co/spaces/JQL-AI/… 🔧 𝐆𝐢𝐭𝐇𝐮𝐛: github.com/JQL-AI/JQL-Ann…

📢 New work published in TMLR: Forecasting Company Fundamentals ✅ 24 ML models (stat/statistical + DL + pretrained) benchmarked ✅ DL models (RNNs, Transformers, etc.) beat classics, esp. for uncertainty ✅ Forecasts rival human analysts (!) A rich playground for applied ML!🌍