Tiktokers will cry at the Corner 😛 #stablediffusion #aivideo #aianimation Bring Life to your AI shorts/Reels and make VIRAL🔥 AI video + IC Lightmap = 🔥 Workflow coming Soon....⏰ Now in ComfyUI🙂 Here is the output 🧐👇

In this technical report, we used MiraData, MiraBench, and MiraDiT to achieve Sora-like consistency. While all three components contributed, we believe the miracle lies in the right way of thinking about the data—hence, MiraData😊⛳️ Project Page: mira-space.github.io

I love that we can use AI to create fan fiction based on movies that haven’t come out yet. Here’s a quick Fantastic Four team-up clip, featuring the new cast, coming in 2025 to the MCU. 🔊🔊

🚀 Join us at our poster session: "Drag Anything: Motion Control for Anything using Entity Representation" 📍 Poster Location: #228 🗓️ Wed, 2nd Oct, 4:30 PM – 6:30 PM CEST Don’t miss it! #AI #ComputerVision #Research #MotionControl #ECCV

Welcome to join Rui's talk!

ICLR 2025 will have another blogpost track! If you have new intuitions on past work, noticed key implementation details for reproducibility, have insights into the societal implications of AI, or an interesting negative result, consider writing and submitting a blogpost.

good consistency for the multi-view.

⚡Say hello to Wonder Animation⚡ - Kickstart your animated project with groundbreaking Video to 3D Scene technology. Convert live footage into 3D scenes with all camera setups, character body and face animation, and fully editable elements in one 3D space. Discover your next

so promising.

ShowUI One Vision-Language-Action Model for GUI Visual Agent

Exciting application! However, when will the open-source community support multi-person LoRA? Our new work, MovieBench (weijiawu.github.io/MovieBench/), can help advance progress in personalized multi-person video modeling. The data will be released within a month—stay tuned!

This performance appears to be remarkably close to that of a real product.

Nike used AI for their new ad. Check out the process in the comments 👇🏻

So powerful

Glad to see GPT-5’s dynamic router balances fast answers and deep reasoning based on query complexity and user intent. In our recent work “Think-or-Not”: arxiv.org/pdf/2505.16854 We study this adaptive reasoning and introduce an easy-to-follow “thought-dropout” mechanism that

Time Is a Feature: Exploiting Temporal Dynamics in Diffusion Language Models "Our work here reveals a critical phenomenon, temporal oscillation, where correct answers often emerge in the middle process, but are overwritten in later denoising steps. To address this issue, we

Check out our latest survey on visual reinforcement learning, and feel free to open a pull request to help keep it growing! Github: github.com/weijiawu/Aweso…

😫Struggle with preparing presentation video before Deadline (such as NeurIPS)? 🔥🔥Thrilled to share our latest work — Paper2Video — automatically generates presentation videos from papers!! 🚀🚀Just provide your paper➕a portrait photo➕a short audio sample — our Paper2Video