PhD student @ Weizmann | Brain–Vision Models, Multimodal AI, Brain Interpretability
Boston
Joined May 2024
- Tweets28
- Following145
- Followers94
- Likes75
Navve wasserman retweeted
Credit assignment in AI is a growing concern. While LM data attribution could be the solution, it's not ready yet.
Good news: @lamsheeper's first work on CATT pushes this a step forward by finding two simple fixes that resolve systematic failures in data attribution!
Can we trace an AI model's answers back to its training data?
Surprisingly, even when we know exactly which documents taught a model a fact, attribution methods often fail to find them.
Our CATT benchmark investigates why and uncovers two simple fixes
catt.baulab.info🧵
Navve wasserman retweeted
How multimodal are unified multimodal models, really?
In our new paper, @benno_krojer and Donghao Zeng show that steering vectors learned in one modality transfer zero-shot to another (text ↔️ image)!
I'm excited about this evidence for shared causal directions across modalities!
New preprint:
Cross-modal steering vectors transfer zero-shot in both directions 🖼️ ↔️ 📝
From text generation to image generation and vice-versa
Exciting evidence for a shared conceptual space!
Fantastic effort led by Donghao Zeng and awesome co-authors @nsubramani23 & @TamarRottShaham
Details 🧵👇
Navve wasserman retweeted
🎬 Excited to share our paper: Joints, Camera, Action! (JCA)
JCA generates video & 3D motion together for any rig!
Video models generate motion you can watch. But what if you could use that motion to animate a 3D asset, with motion that responds to its surroundings?
research.nvidia.com/labs/par…
🧵
Navve wasserman retweeted
My favorite artist has always been M.C. Escher. His "Print Gallery" has been hanging in my home for years.
Recently I bumped into a 3B1B video about it that was really awesome and I thought it would be cool to generate such art. A bit out of my comfort zone, but luckily @Sophie0tq took the lead (🧵).
Navve wasserman retweeted
TMLR has faced a deluge of submissions, necessitating stricter desk rejection policies due to limited reviewer capacity
Co-EiC Nihar Shah reached out to authors of 10 papers slated for desk reject. Could they answer questions about their *own* submission?
medium.com/@TmlrOrg/asking-a…
Navve wasserman retweeted
Our full schedule is live! curateddata.github.io/schedu…
Come stop by our full-day #ECCV2026 workshop this Wednesday, located in Malmö Arena Panorama 2, to hear from our great lineup of speakers!
@eccvconf
Navve wasserman retweeted
10 days to submit your work to DevAI @ NeurIPS 2026!
Details & submissions:
sites.google.com/view/devai-…
Navve wasserman retweeted
How similar are two images? Prior metrics (e.g., LPIPS, DreamSim) give just a single score. But actually, there are multiple *senses* of similarity (color, pose, etc.)
We introduce TPIPS -- Text-Prompted Image Perceptual Similarity
"pip install tpips"
peterwang512.github.io/TPIPS 🧵
Navve wasserman retweeted
📢 Call for Papers!
Excited to announce DevAI: Developmental Perspectives on AI, a new workshop at #NeurIPS2026!
We invite papers, datasets, and benchmarks at the intersection of AI and human development.
📅 Archival: Aug 29
📅 Non-archival: Oct 13
🔗 sites.google.com/view/devai-…
Organizing committee: @MengmiZhang, @KelseyRAllen, @YehonatanAv, @shimonUllman & @ShifyTreger
#DevelopmentalAI #MachineLearning #CognitiveScience #Neuroscience #AIResearch
🤖 Made with AI
Navve wasserman retweeted
We're hosting the workshop on Interpretability for Discovery🔬 at NeurIPS 2026, Atlanta!
AI models predict protein folding, forecast weather patterns, and more. They might be learning new patterns we haven't figured out yet. Can we use interpretability to read out these patterns?
Navve wasserman retweeted
Can we tell whether data domains cooperate or compete during pretraining?
Adding code to the mix makes models better at math, while some other combinations hurt each other. We call this data synergy. Turns out you can incorporate data synergy into scaling laws and estimate it 🧵
Navve wasserman retweeted
Do reasoning models internally represent abstract properties of their own chain of thought (such as "which steps are important"), while not surfacing these properties in their tokens?
Navve wasserman retweeted
How can we tell whether a brain region causally represents a visual concept, rather than merely correlating with it?
Introducing BrainCause, a framework combining generative and brain models to create controlled stimuli and causally test neural representations.
More below 🧠👇
How can we tell whether a brain region causally represents a visual concept, rather than merely correlating with it?
Introducing BrainCause, a framework combining generative and brain models to create controlled stimuli and causally test neural representations.
More below 🧠👇
6️⃣ Are these representations consistent across people?
Despite individual differences in cortical anatomy, BrainCause identifies similar concept-selective regions across multiple subjects, providing evidence that the discovered organization is reproducible.
7️⃣ 📄 Paper:
arxiv.org/abs/2605.23895
🤗 Dataset:
huggingface.co/datasets/Brai…
🌐 Project page:
yuvalgol123.github.io/BrainC…
This work was done in collaboration with @GolbariYuv39835, Matias Cosarinsky, @RBeliy95358, @AudeOliva, Antonio Torralba, Michal Irani, and @TamarRottShaham
