The one true benchmark for personalization, results from Pika are crazy good!

Very cool idea: build a personalization encoder so that each query selects the right values encoded from the input image, resulting in a more expressive representation. The results look awesome!

VideoJAM is our new framework for improved motion generation from AI at Meta We show that video generators struggle with motion because the training objective favors appearance over dynamics. VideoJAM directly adresses this **without any extra data or scaling** 👇🧵

Thrilled to share that our work has been accepted to #ICLR2025 !🎉📜 This research introduces SoTA attribution methods for Mamba and other gated RNNs (such as Griffin and RWKV) across both vision and NLP. The image captures one of the most exciting insights from our work! [1/3]

Turns out diffusion models have instance-level representations, and we use that to make them count right! Excited to share that Make It Count has been accepted to #CVPR2025! 🎉 🔗 Check it out: make-it-count-paper.github.io

🚀 Excited to share OmnimatteZero: Training-Free Real-Time Omnimatte with Video Diffusion Models! 📄 Paper: arxiv.org/abs/2503.18033 🌐 Project: dvirsamuel.github.io/omnimattezero.… 🧵👇

Are you at #NAACL? Come to Ballroom C at 10:15 for our talk on Padding Tone 🐻 We show that text-to-image models don’t just ignore padding tokens—they actively use them during image generation, depending on the model’s architecture and training. Project page in the first comment.

📄🚨New! Attribution methods that assign relevance to tokens are key to extracting explanations from Transformers and LLMs. Yet, all existing approaches produce fragmented and unstructured heatmaps (see image)! Why does this happen? and how can we fix this chronic issue? See🧵1/5

1/6🚀 New #ACL2025Findings: We show you can predict if Chain-of-Thought (CoT) reasoning will succeed — before any tokens are generated! This works with LLMs not specifically trained for reasoning—meaning powerful signals emerge naturally in early processing.

Tired of manual #ComfyUI workflow design? While recent methods predict them, our new paper, FlowRL, introduces a Reinforcement Learning framework that learns to generate complex, novel workflows for you! paper [arxiv.org/abs/2505.21478]

🎉 Thank you AK for featuring our work! We’re happy to release the 💻 code & 🤗 Hugging Face demo for Add-it, our #ICLR2025 paper: 💻 Code: github.com/NVlabs/addit 🤗 Demo: huggingface.co/spaces/nvidia/…

BTW, I just tried out Anycoder by AK and it's awesome! I simply gave it the Add-it paper PDF and a link to our code, and it instantly generated an academic project page. It already supports Kimi K2 and many other models! Try it out 👉 huggingface.co/spaces/akhaliq…

OmniSVG demo & weights finally dropped on Hugging Face🔥🔥 ✨ end-to-end multimodal SVG generator ✨ leverages pre-trained VLMs ✨ works all the way from simple icons to intricate anime characters