Heading to #CVPR2025 in Seattle next week? Don't miss our open-bar happy hour where you can connect with us and other top minds in the field. Capacity is limited, so RSVP today: 🔗lu.ma/ul994esr #CVPR2024

🏆We are thrilled that our Embodied Agent Interface (x.com/manlingli_/sta…) received the Best Paper Award at SoCal NLP 2024 (~top 0.4%)! ❤️A big thank you to our incredible team Jiajun Wu Fei-Fei Li Percy Liang Jiayuan Mao Ruohan Zhang Weiyu Liu Shiyu Zhao Qineng Wang

I'm building a new research lab Engineering Dept focusing on 4D computer vision and generative models. Interested in joining us as a PhD student? Apply to the Engineering program by Dec 3 🗓️ postgraduate.study.cam.ac.uk/courses/direct… ChatGPT's "portrait of my current life"👇 elliottwu.com

1/ [NeurIPS D&B] Introducing HourVideo: A benchmark for hour-long video-language understanding!🚀 500 egocentric videos, 18 total tasks & ~13k questions! Performance: GPT-4➡️25.7% Gemini 1.5 Pro➡️37.3% Humans➡️85.0% We highlight a significant gap in multimodal capabilities🧵👇

#NeurlPS I’ll be in Vancouver from December 8-16 and presenting IKEA Manuals at Work . 📅Wednesday, Dec11,4:30-7:30 PM PST 📌West Ballroom A-D(#5207) Let's grab a coffee ☕️or have a chat—I'm also exploring Fall 2025 PhD opportunities. Would love to connect!! :> #NeurlPS2024

#NeurIPS2024 We will present how to transfer procedural planning from 2D to 3D space. Yunong Liu is amazing to lead this large project efforts and is looking for PhD positions. Reach out to her if you are interested! 📅Wednesday, Dec11, 4:30-7:30 PM PST 🗺️West Ballroom

Thanks much Fei-Fei Li! We will present Embodied Agent Interface in the oral session right after Fei-Fei Li’s keynote. Come and join us! #NeurIPS2024 📆Oral: Wed 11 Dec 3:50-4:10pm PST, Oral Session 2A Agents 📆Poster: Wed 11 Dec 4:30-7:30 pm PST, West Ballroom A-D #5402

#NeurIPS2024 Grateful for all who visited my poster 😍 excited to meet with new & old friends ! Loved presenting our work with Cristóbal Eyzaguirre Ercilla & Juan Carlos Niebles. Small hiccup when my website QR got deactivated by a third-party converter pushing subscriptions 😅 but couldn't dampen

🚀 RAG vs. Long-Context LLMs: The Real Battle ⚔️ 🤯Turns out, simple-to-build RAG can match million-dollar long-context LLMs (LC LLMs) on most existing benchmarks. 🤡So, do we even need long-context models? YES. Because today’s benchmarks are flawed: ⛳ Too Simple –

🤖 Ever wondered what robots need to truly help humans around the house? 🏡 Introducing 𝗕𝗘𝗛𝗔𝗩𝗜𝗢𝗥 𝗥𝗼𝗯𝗼𝘁 𝗦𝘂𝗶𝘁𝗲 (𝗕𝗥𝗦)—a comprehensive framework for mastering mobile whole-body manipulation across diverse household tasks! 🧹🫧 From taking out the trash to

Introducing Camera Angle Concepts for #Ray2 — a new way to control your POV. Choose your angle and consistently frame your story with cinematic perspectives like overhead, selfie, low angle, over the shoulder, aerial, and more. Available now in #DreamMachine.

Ray2 is coming to #AdobeFirefly — bringing the next frontier of video generation to Firefly and Firefly Boards. From simple, natural prompts to cinematic motion, create with #Ray2 wherever Firefly is available. lumalabs.ai/blog/news/ray2…

We are very excited announcing our MLL lab! We are looking for collaborators on RAGEN, VAGEN, Chain-of-experts, T*, LongVideoHaystack, foundation models for embodied agents, etc mll-lab-nu.github.io

🚨 Super excited to share that our paper "GSM-Infinite: How Do Your LLMs Behave over Infinitely Increasing Context Length and Reasoning Complexity?" got accepted at ICML 2025! 🧠📏🎉 P.S. 😅 Struggling to make your long-context LLM submission stand out at NeurIPS? 🧠 Give

Reframe is now available to all subscribers.

Introducing Modify Video. Reimagine any video. Shoot it in post with director-grade control over style, character, and setting. Restyle expressive performances, swap entire worlds, or redesign the frame to your vision. Shoot once. Shape infinitely.

Reimagine the video in anyway you want 🤩 Let’s gooooo 🎬

1/ Model architectures have been mostly treated as fixed post-training. 🌱 Introducing Grafting: A new way to edit pretrained diffusion transformers, allowing us to customize architectural designs on a small compute budget. 🌎 grafting.stanford.edu Co-led with Michael Poli

This is insane🤣