🔥Telegram İfşa

We’re thrilled to announce the launch of the MolSS Reading Group! 🚀 🔬 MolSS = Machine Learning for Molecular Simulations and Sampling 🔥 First session: June 3rd, 3 PM (UK time) by Jose Miguel Hernández-Lobato 🗓️ Regular sessions: Every other Tuesday at 4 PM (UK time), 45–60 minutes

Interested? Please check out the paper and code below for more! Work with Pasquale Minervini is hiring postdocs! 🚀 Edoardo Ponti and antonio vergari ⚔️ not at #ICLR2025 Arxiv: arxiv.org/abs/2505.13138 Code: github.com/HEmile/neurosy…

6. Neurosymbolic Diffusion Models 🔑 Keywords: Neuorsymbolic, diffusion models, dependencies, uncertainty quantification, visual reasoning 💡 Category: Multi-Modal Learning 🌟 Research Objective: - To address the limitations of standard Neurosymbolic (NeSy) predictors in

If you're an international PhD student at Harvard studying computer vision and your visa is cancelled, reach out to me or others in Europe. Don't despair. I'm sure we can find you a great place to carry on your research.

Check out SpARE, an extremely effective method for steering LLMs by editing their SAE activations! arxiv.org/abs/2410.15999 (NAACL’25 Oral)

meanwhile on the London tube.. TfL

🗓️ Deadline extended: 💥2nd June 2025!💥 We are looking forward to your works on: 🔌 #circuits and #tensor #networks 🕸️ ⏳ normalizing #flows 💨 ⚖️ scaling #NeSy #AI 🦕 🚅 fast and #reliable inference 🔍 ...& more! please share 🙏

I spent this morning reproducing with o3 Anthropic's result that Claude Sonnet 4 will, under sufficiently extreme circumstances, escalate to calling the cops on you. o3 will too: chatgpt.com/share/68320ee0…. But honestly, I think o3 and Claude are handling this scenario correctly.

“You must never be fearful about what you are doing when it is right.” -- Rosa Parks

The Bitter Lesson strikes again

So how well do the best VLMs (e.g. Gemini 2.5 Pro, GPT-4o, Claude 3.7) perform on VideoGameBench? 🥁 Really bad! Most models can’t progress at all in any games on VideoGameBench, which span a wide range of genres like platformers, FPS, RTS, RPGs, and more!

During my PhD I have actually developed a real love for presenting! Not what I expected 😂 I wanted to share my top ten tips about presenting that I learnt as a secondary school teacher many years ago. Hope they're helpful, and I would love to get your thoughts!🙂 1. Relax and

As promised, we put on Arxiv the proof we did with Gemini. arxiv.org/pdf/2505.20219 This shows that the Polyak stepsize not only will not reach the optimum, but it can cycle, when used without the knowledge of f*. Gemini failed when prompted directly ("Find an example where the

I might've gotten a little too excited designing our #CVPR poster on model merging… Come find it in Nashville 🎸 I promise it won’t be hard to spot! Paper 👉 arxiv.org/abs/2412.00081 Joint w/ Antonio A. Gargiulo, Maria Sofia Bucarelli, Simone Scardapane, Fabrizio Silvestri, Emanuele Rodolà

Excited to present PLAYPEN, an environment for learning through dialogue game self-play. Are you interested in LLM post-training techniques and how to make them better "language users"? Read this thread! #GenAI #NLProc #HRI #AI #ML

We've got a major update to our preprint on adversarial regret guarantees for Thompson sampling! As before, I think this is one of the most important projects I've worked on due to new algorithmic primitives that it - in principle - unlocks. Thread below on what's new!

If anyone has done any work improving the robustness of NLI models and we didn't cite you in our appendix, please share a link to your work - I would love to include it Our appendix related work on NLI robustness is a bit of a monster 😅🧌 x.com/_joestacey_/st…