#VideoReason We are open-sourcing the entire VBVR stack to speed-up the arrival of video reasoning as the next fundamental paradigm of intelligence
- 150+ synthetic generators
- 1 million training clips
- Cloud-scale data factory
- Unified EvalKit
- 100 rule-based evaluators
- Strong baseline model
Checkout at https://t.co/lOtJzJYC52
Welcome to Season 5!
From Tolman’s cognitive maps to Sherrington’s neural circuits, @DengHokin will re-examine the emergence of latent representations in biological minds.
🕘 Oct 23 2025
🇸🇬SGT – 21:00 (UTC+8)
🇺🇸EST – 09:00 (UTC-4)
🔗https://t.co/IuM33VtvIT for talk details!
💥We are excited to share that our #Video2Animation feature is now live at @MyolabAI's discord servers.
We're giving away massive 𝐟𝐫𝐞𝐞 credits for early users. Head now to 👉 https://t.co/c4rLQxorBM
#embodied All forms of biological intelligence are grounded movements🏃♂️ muscles & motor neurons 🧠 emerge before visual cortex & rods & cones in eyes 👁️
Building monocular better-than-mocap-studio #video2motion is our critical step towards human embodied intelligence.
🔥 Huge thanks to @ylecun and everyone for reposting our #ICML2025 work! 🚀
✨12 core abilities, 📚1503 tasks, 🤖230 MLLMs, 🗨️11 prompts, 📊2503 data points.
🧠 We try to answer the question:
🔍 Do Multi-modal Large Language Models have grounded perception and reasoning?
🚀 Dive in 👇
🌐 Project: https://t.co/SyO2lunp7q
📄 Paper: https://t.co/Po5pS14iWz
📝 OpenReview: https://t.co/ipBXg9oSiC
📊 Dataset: https://t.co/AwWbeCw7CQ
💻 Code: https://t.co/QxPm5MJqU7
😎 Team: https://t.co/ZdJsDnB1OF
💥 Huge thanks to my amazing collaborators @Williamiumli, @Kaiagaoqy, Icy Wang, Tianwei Zhao, @HaoranSunMr, Zoey Lyu, Robert Hawkins, Nuno Vasconcelos, @TalGolanNeuro & @carrot0817_ 💐
🙌 Super grateful to @fredahshi, @ziqiao_ma, @ZimingLiu11, @_beenkim, @TomerUllman & @gershbrain for the 🔥 discussions and wisdom 🧠
💡 Massive shout-out to @_emliu, @zory_zhang, @Pinyuan3, and everyone in the @GrowAiLikeChild community for their constant support & belief ❤️🚀
7) The result is astonishing 😱 All models - either they become totally illusory or they become completely short-cutted. None of the models is even close to core knowledge.
No models are able to get both the manipulation tasks and the control tasks right at the same time, which seems to completely trivial to humans ... (11/n)
#ICML#cognition#GrowAI We spent 2 years carefully curated every single experiment (i.e. object permanence, A-not-B task, visual cliff task) in this dataset (total: 1503 classic experiments spanning 12 core cognitive concepts).
We spent another year to get 230 MLLMs evaluated and got in total 3800K inference data points with 11 different prompt formats.
Nonetheless, cognitive psychologists first spent 100 years to uncover and illuminate the developmental trajectory of human cognition.
Here we stand at a pivotal moment—learning human developmental psychology on the shoulders of giants but witnessing the rapid rise of a new form of intelligence.
A thread (1/n)
I love the academic's excuse "I can't make it, I'm 'travelling'." That could mean anything from presenting cutting-edge research at a conference to experiencing ego death at an ayahuasca retreat
Announcing the keynote speakers for #ICLR2025! Speakers will cover topics ranging from foundational advances in language models, AI safety, open-ended learning, and the nature of intelligence itself. https://t.co/nL73E3KtbG