Noam Rotstein retweeted
Coding agents can build an app. What if they could give it a body?
Meet Metatron: a versatile System 1 controller for humanoids. A goal-based interface connects System 2 planners, including coding agents, to whole-body movement.
A few examples below. 1/5
Noam Rotstein retweeted
1/4 We trained an 8B safety classifier on just 1,275 carefully annotated prompts.
It beat GPT-5.4, Claude Sonnet 4.6, and leading open-weight guardrails trained on far more data, on average across 5 safety benchmarks.
Quality > quantity.
Noam Rotstein retweeted
Do vision-language models look at the image regions they describe? In our new paper, we present Self-Saliency, a method for aligning visual attention with reasoning text.
research.nvidia.com/labs/par… w/ @GalChechik @DalalGal
Noam Rotstein retweeted
Excited to share our work, “Token-to-Token Alignment of Text Embeddings for Semantic Blending,” accepted as a NeurIPS 2026 Spotlight 🌟
Can we directly interpolate between semantic concepts?
Text embedding spaces already encode a continuous semantic structure, the missing piece is alignment.
Once aligned, simple interpolation becomes meaningful, naturally enabling multiple applications.
Joint work with @MokadyRon @OPatashnik @DanielCohenOr1
Noam Rotstein retweeted
Meet DyRAD: Radar NVS for dynamic driving scenes.
📡🚗 Radar directly senses motion through Doppler, but prior radar NVS methods ignore this signal where it matters most: dynamic scenes.
🌐 dyrad-nvs.github.io
📄 arxiv.org/abs/2609.39841
💻 github.com/Dyrad-NVS/DyRAD
🧵
Noam Rotstein retweeted
1/11 Can a stronger coding model rescue a run started by a cheaper one?
Can a cheaper model continue a stronger model's work - retaining quality at lower cost?
The two directions behave very differently...
Check out our new paper: The Handoff Tax 🧵
📄 arxiv.org/abs/2608.24358
Noam Rotstein retweeted
🥳 Excited to share UniProbe: A learnable, token-level hallucination detection for large VLMs 🔎
UniProbe looks inside a VLM’s internals to identify hallucinated response tokens and mitigate them before they reach the user.
📉 55% fewer hallucinations
⚡ Just 1.06× latency
Noam Rotstein retweeted
I'm going to present Fast-AR at #ICML26 tomorrow (Tuesday) @ 2pm - poster #1204!
Come and say hi 👋
If you're around this week, feel free to DM me.
Project page: dvirsamuel.github.io/fast-au…
Details below ⬇️
Noam Rotstein retweeted
Excited to share that my first PhD paper, RadarGen, got accepted to #ECCV2026! 🎉
Let’s talk about radar simulation in Malmö 🇸🇪
📢 RadarGen: Automotive Radar Point Cloud Generation from Cameras
Can we generate realistic radar point clouds solely from camera images? 🚗📡
We introduce RadarGen, a diffusion-based framework that synthesizes radar returns aligned with visual scenes.
radargen.github.io
Noam Rotstein retweeted
Glad to share the main research of my master's degree, VideoMDM! Please go check it out! videomdm.github.io
Excited to share our new paper: VideoMDM 📢
We propose a principled framework for training 3D motion diffusion models (e.g. MDM), using only 2D supervision from monocular videos -- no 3D ground truth required.
Project: videomdm.github.io
Paper: arxiv.org/abs/2606.13364
🧵
Noam Rotstein retweeted
A new technology developed at the Technion could significantly expand access to AI-powered video creation
Called Time to Move (TTM), the system enables users to create and precisely control realistic AI-generated video clips using simple mouse movements. Unlike previous approaches, TTM requires no retraining, no additional computational cost, and can be integrated directly into existing video-generation models.
Developed by Dr. @orlitany , Prof. Ron Kimmel, and students Asaf Singer, Noam Rotstein, and Amir Mann from the Henry and Marilyn Taub Faculty of Computer Science, TTM was presented at @iclr_conf 2026, one of the world's premier conferences in deep learning and artificial intelligence.
Noam Rotstein retweeted
🎉 Happy to share our latest work: Bootstrap Your Generator: Unpaired Visual Editing with Flow Matching (accepted to #ICML2026)!
TL;DR: We train image and video editing models without any paired data.
No ground-truth edit data, no external reward models.
Noam Rotstein retweeted
1/7
When rewards conflict, what should RL post-training of diffusion models optimize?
In visual generation, objectives are often in tension:
Prompt adherence can conflict with source preservation.
Photorealism can conflict with stylization.
In our new paper, ParetoSlider, we introduce a multi-objective RL framework that trains a single diffusion model for continuous control over competing reward objectives 🧵
Noam Rotstein retweeted
Tired of 3D generations that look synthetic? They don’t have to.
Excited to share that Realiz3D is accepted to #CVPR2026 🎉
A framework for training diffusion models that are 3D-consistent, controllable, and photorealistic.
🔗 Project page: idosobol.github.io/realiz3d/
🎥 Teaser ↓
Noam Rotstein retweeted
Excited to present DRoPS! A novel method for dynamic 3D reconstruction from monocular videos, accompanied by a static-prescan of the dynamic object.
🌐 Project: drops-dynamics.github.io
📄 Paper: arxiv.org/abs/2603.24770
🧵 [1/7]
Noam Rotstein retweeted
Excited to share that our work, “Image Generation from Contextually Contradictory Prompts,” was accepted to CVPR 2026 🎉
Diffusion models fail on some seemingly simple prompts.
Why do they ignore what you asked for?
We show why, and how to fix it with a simple, training-free method.
Joint work with @OPatashnik @OmerDahary @MokadyRon @DanielCohenOr1
Noam Rotstein retweeted
SemanticMoments - Semantic motion similarity
How do you find videos with similar motion?
It’s harder than it sounds.
Models like VideoMAE and V-JEPA encode motion, but their embeddings are dominated by appearance.
So how do we build a compact embedding for motion similarity?
Joint work with @kfir99 @OPatashnik @BenaimSagie @MokadyRon
Noam Rotstein retweeted
The potential of this tech is just scratching the surface!
Watch this Blender layout transform into a final render using my Time-to-Move (TTM) workflow:
Noam Rotstein retweeted
I wanted to try out @mickmumpitz ComfyUI workflow that lets you animate movement by manually shifting objects or images in the scene. Link to his tutorial below 👇
Noam Rotstein retweeted
This just in: Lumana is now the world’s fastest-growing AI video security company!
50,000 cameras surpassed in just over a year since emerging from stealth.
Read more: bit.ly/3XS50u2
#AIVideoSecurity #AI