Nir Goren (@nirgoren) profile picture

Researching AI https://nitter.gq/t.co/eMCNT73s4T

Joined May 2018
Photo by Nir Goren (@nirgoren)Photo by Nir Goren (@nirgoren)Photo by Nir Goren (@nirgoren)Photo by Nir Goren (@nirgoren)Photo by Nir Goren (@nirgoren)Photo by Nir Goren (@nirgoren)
Nir Goren retweeted
Can you watermark any generative model just by choosing what to release? Our black-box method is provably distortion-free and robust to content-preserving edits. With @OPatashnik and @OREAXEAX Paper → arxiv.org/abs/2611.23208 Project page → avivyossef.github.io/distort…
7
10
1
38
1,746
1/7 ViTs give us powerful semantic features, but coarse grids and spatial artifacts hurt dense prediction. Introducing ElasticDino (WACV 2027): a lightweight adapter that produces sharp, pixel-aligned DINOv2/v3 features by moving features rather than rewriting them.
1
5
1
17
265
Multi-turn image editing is notoriously challenging: as edits accumulate, unintended changes and artifacts build up. Not anymore!😁 Introducing ACEdit ♠️ A training approach that counteracts accumulated drift across successive image edits, preserving scene details as the changes add up. 🌐 saradorfman1.github.io/ACEdi… 1/5
6
23
5
89
9,040
[1/8] 🚀 New paper: APEX: Asynchronous Prefix Denoising for Extreme Compression We give the same token multiple diffusion timesteps at once. APEX delivers better generation at the same compute with 4× fewer tokens. 🧵 🌐 Project page + code: daviddinkevich.github.io/APE…
3
19
2
113
5,921
Nir Goren retweeted
Glad to share our work, IntrinSync! We jointly decomposes an image into editable intrinsic maps and renders them back to RGB, with cycle consistency to keep the maps coherent and support physically grounded image editing in a unified framework. arxiv.org/abs/2610.11138
3
2
18
2,423
Nir Goren retweeted
When generating an image with an MM-DiT, the text tokens absorb information from the evolving image through multimodal attention. But what do these contextual tokens actually capture? In our new work, “Learning to Read,” we show that contextual tokens can be read as plain text, revealing surprisingly detailed visual semantics. Building on this finding, we introduce a new way to train diffusion models: supervising their contextual tokens to improve generation quality and distributional coverage. 1/n
10
25
1
106
4,367
You ask for a talking parrot, but the model makes the pirate talk AND gives him a beak! 🦜😮‍💨 -> Introducing The Attention Triangle in Audio-Video Models: how semantics leak between text, audio, and video, and how we steer them back. 🧵 [1/5]
3
27
1
60
1,988
Nir Goren retweeted
1/ New paper: Parallel Decoding Distillation (PDD)!🚀 To accelerate diffusion/flow models, trajectory distillation usually merges consecutive ODE steps into 1 larger step. PDD changes this by predicting MULTIPLE ODE steps in 1 network eval.🧵 w/ @julberner @ArashVahdat Chao Liu
5
38
5
169
24,686
🔊 Audio on for this one! Code + checkpoint for ScenA, our multi-speaker audio scene generation model, are out! Reference voices + a prompt → a full conversational scene, prompt alone decides who speaks where. github.com/finmickey/scena arxiv.org/abs/2606.19325
3
14
57
5,374
🌟🚀 Excited to share our latest work: "Keep The Essentials: Efficient Reference Conditioned Generation via Token Dropping"! TL;DR: Stop wasting compute on redundant tokens! We introduce SparseContext that drops reference tokens for speeding up reference-based image generation⚡
3
14
1
51
10,181
Generating images with AI is fun 🤓 Finding the image? Not so much… The first result is rarely the one. Or the second. Or the third. 🥵 So we built a way to generate diverse possibilities along meaningful semantic axes 🧭 and organize them into a gallery you can effortlessly browse 🧚🏻 Introducing our #ECCV2026 paper: Semantic Browsing: Controllable Diversity for Image Generation ✨ 1/5
2
17
3
37
2,585
Nir Goren retweeted
Can a small academic team build a strong text-to-image model using only public datasets? Introducing i1: a simple, fully open recipe for strong text-to-image models
11
78
9
442
71,732
1/7 New paper 🧵 ScenA generates a multi-speaker audio scene: overlapping speech, laughter, real room noise, from a text description and a few reference voices. When trained in the obvious way, it ignores the text and decides who speaks on its own. We found out why and fixed it.
1
22
4
57
6,220
FUN to share that Sync-LoRA has been accepted to #ECCV2026! 🤗✨🎬 the field moves fast, but the core seems to remain: 1️⃣ IC-LoRA enables diverse edits 2️⃣ high-quality data is key (even if it's a small amount!) Huge thanks to @arash_mham @OPatashnik @DanielCohenOr1
[1/4] Sync about it… 💭✨ Editing a portrait video yet keeping it fully synced with the original across the entire sequence. Read more about Sync-LoRA: sagipolaczek.github.io/Sync-… 🚀
3
7
1
35
14,398
Nir Goren retweeted
Two months ago, I vaguely posted a number: 0.9 FID, one-step, pixel space. Now it is 0.75, and can be even lower. Many wonder how. I thought it might end as a small FID prank: simple and deliberate. It started with one question: can FID be optimized directly, and what does it reveal? Introducing FD-loss.
66
159
41
997
265,051
Nir Goren retweeted
Excited to share our work accepted to #SIGGRAPH2026 ! Video generation models struggle with something few talk about: their transformations don't evolve smoothly. You get long boring stretches... then a sudden semantic jump where everything "catches up" at once.
7
27
1
71
6,057
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 🧵
4
28
4
87
14,459
We developed a simple, sample-efficient online RL technique for post-training image generation models. We see it as a possible steerable alternative to CFG, driven by any scalar reward, including human preference.
11
42
8
379
71,719
Nir Goren retweeted
[1/5] Is Text Enough for Control? 🐇 Text-driven video editing lets you describe *what* to change. But what about *how much*? We introduce Adaptive-Origin Guidance (AdaOr). A joint work with @DecartAI and @TelAvivUni 🧪 accepted to #SIGGRAPH2026.
6
24
1
67
12,056
Nir Goren retweeted
Video models as Physics simulators. 🌍🎥 [1/] In our latest work, WinDiNet, we finetuned a pre-trained video model into a differentiable physics engine. 1000x faster than traditional CFD solvers. Project page: rbischof.github.io/windinet_… Abs: arxiv.org/abs/2603.21210
1
37
1
210
14,265