🔥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

Looking forward to speaking at this workshop on Spurious Correlationa and Shortcut Learning at ICLR 2025! I was very into this line of work towards the end of my PhD. I still have a lot of curiosity about this, and think there are some significant implications for AI safety.

🎉 The holidays are over—get ready to share your groundbreaking research! 🚨 Submit to the SCSL Workshop: 1️⃣ New spurious corr. benchmarks for various domains (incl. multimodal). 2️⃣ Robustification methods to tackle spurious corr. & shortcut learning. 3️⃣ Foundations of SCSL. /1

Pragmatically, we can say that AGI is reached when it's no longer easy to come up with problems that regular people can solve (with no prior training) and that are infeasible for AI models. Right now it's still easy to come up with such problems, so we don't have AGI.

I am very excited about this new paper and more generally this line of work. See Ali's detailed description of how adding this memory unit can enable transformers scale to much longer context length.

Looking forward to see insightful work submitted to this workshop.

🔥🙏🏼 #AROS 💍 is accepted to #ICLR2025 ICLR 2026 ! So proud of HOSSEIN - what an awesome way to kick off his first grad school project 👌👌 Check out the updated arXiv version of the paper, open code (including python package) below ⬇️ #AROS💍 leverages neural ODEs and

We are honored to host Prof. Sutton in this week Wednesday at Sharif. Please sign up if you’re interested: docs.google.com/forms/d/e/1FAI…

We are delighted to host Michael this week in our RL course! Looking forward to his presentation.

Richard Sutton, father of Reinforcement Learning, just gave a great talk where he said "decentralized collaboration is the source of all human flourishing and the source of all that's good in the world." And if you're in favor of this, "the enemy" are the people calling for

🚀 SCSL Workshop at ICLR 2025 (Singapore, April 28) is seeking sponsors! Join us in tackling spurious correlations & shortcut learning with top researchers from NYU, UdeM, UCLA, UMD, SUT, and Google DeepMind. Support cutting-edge AI research & gain visibility! 🔗 Learn more:

⚡️Really thrilled that #textgrad is published in @nature today!⚡️ We present a general method for genAI to self-improve via our new *calculus of text*. We show how this optimizes agents🤖, molecules🧬, code🖥️, treatments💊, non-differentiable systems🤯 + more!

🪂Understanding R1-Zero-Like Training: A Critical Perspective * DeepSeek-V3-Base already exhibits "Aha moment" before RL-tuning?? * The ever-increasing output length in RL-tuning might be due to a BIAS in GRPO?? * Getting GRPO Done Right, we achieve a 7B AIME sota! 🧵 📜Full

🔥 How do you build a state-of-the-art Vision-Language Model with direct RL? We’re excited to introduce VL-Rethinker, a new paradigm for multimodal reasoning trained directly with Reinforcement Learning. 📈 It sets new SOTA on key math+vision benchmarks: - MathVista: 80.3 → 🥇

🚨 Join us at the Workshop on Spurious Correlation & Shortcut Learning (SCSL) at #ICLR2025! ICLR 2026 🗓️ April 28, 2025 📍 Garnet 214-215, Singapore EXPO 🌐 More info: scslworkshop.github.io #ICLR2025