Timna

[Ariel Goldstein @GoldsteinYAriel]

New paper! 📷
The brain's Multiple Demand (MD) network ramps up when tasks get hard, so its activity gets read as "effort", while meaning is left to the language network. But when you're following a tough lecture, is MD just working harder, or is it tracking the content?

[Daria Lioubashevski @DariaLioub]

🚨 New preprint!
One idea, many ways to say it – does your brain track those options before you speak?
Using LLMs, we put this to the test:
https://t.co/rCRvuPwDtT
We show for the 1st time that the brain represents many alternatives simultaneously in both listening & speaking 🧵

[Timna @timnaWK]

@EliahuHorwitz @NeurIPSConf Congrats!!

[Timna @timnaWK]

@NoySternlicht @ArielGera2 @RoyBarHaim @Hoper_Tom @noamslonim Congrats! Super interesting work!

[Miriam Havin @MiriamHavin]

Can (A)I change your mind?
New study finds LLMs can be as persuasive as humans — even in real-world, ecological conversations on controversial topics.
https://t.co/BaiPene1M5
#AI #LLM #Persuasion #CogSci2025 @cogsci_soc @timnaWK @GoldsteinYAriel @yanivdover @morankor

[Timna @timnaWK]

@p3ery @AaronBergman18 @GoodfireAI Where can I find the episode?

[Daria Lioubashevski @DariaLioub]

📢Paper release📢
What computation is the Transformer performing in the layers after the top-1 becomes fixed (a so called "saturation event")? We show that the next highest-ranked tokens also undergo saturation *in order* of their ranking.
Preprint: https://t.co/ovwI9JpiAQ
1/4

[Eliahu Horwitz @EliahuHorwitz]

🚨New Dataset Distillation paper 🚨
We propose PoDD, a new dataset distillation setting for a tiny🤏under 1 image-per-class (IPC) budget❗In this CIFAR100 example, current SoTA is 35.5% acc using ~100k pixels, PoDD gets 35.7% with ~40k pixels
Project: https://t.co/8ZI2qBI7bl
🧵👇

[Yuval Reif @ COLM2026 @YuvalReif]

Is dataset debiasing the right path to robust models?

In our work, “Fighting Bias with Bias”, we argue that in order to promote model robustness, we should in fact amplify biases in training sets.

w/ @royschwartzNLP
In #ACL2023NLP Findings
Paper: https://t.co/zYP6FuVpe9
🧵👇

[Asaf Yehudai @AsafYehudai]

Looking for a new SOTA few-shot classifier?

We present QAID!
QAID uses batch contrastive learning with BERTScore &
classifies by retrieving the intent name.

SOTA on few-shot Intent Detection! 🎉
https://t.co/bMxNHwO0gc

Accepted to @iclr_conf 2023 🥳
#NLProc #NLP #ML

[AK @_akhaliq]

Breaking Common Sense: WHOOPS! A Vision-and-Language Benchmark of Synthetic and Compositional Images

abs: https://t.co/HILwgrg907
project page: https://t.co/SRhgnwsD92
@huggingface dataset: https://t.co/EMSxn1dMFT
@gradio demo: https://t.co/iS03zrnDIn

[AK @_akhaliq]

Word-As-Image for Semantic Typography

abs: https://t.co/WZJcC54oBG
project page: https://t.co/wGc6p7cpPZ

[Timna @timnaWK]

@EMostaque Dreamix, editing videos with diffusion models
https://t.co/SIKzDGp7pA

[bleedingedge.ai @bleedingedgeai]

Google announces Dreamix: a model that generates videos when given:
- video + prompt (Video editing)
- input images + prompt (Subject Driven Generation)
- input image + prompt (Image-toVideo

[Aran Komatsuzaki @arankomatsuzaki]

Dreamix: Video Diffusion Models are General Video Editors

proj: https://t.co/rOgXaD3o4P
abs: https://t.co/xwgAxHYV6t

[AK @_akhaliq]

Dreamix: Video Diffusion Models are General Video Editors

abs: https://t.co/It6BuYHU8c
project page: https://t.co/Tks6bU4kmb

present diffusion-based method that is able to perform text-based motion and appearance editing of general videos

[AK @_akhaliq]

GitHub: https://t.co/PCHiRh4qaf

[AK @_akhaliq]

Conffusion: Confidence Intervals for Diffusion Models
abs: https://t.co/LJ9r2K5Ilz
project page: https://t.co/UtM0nyYZqg

[Eliahu Horwitz @EliahuHorwitz]

Happy to share our paper "DeepSIM: Image Shape Manipulation from a Single Augmented Training Sample" was accepted as an 𝗼𝗿𝗮𝗹 presentation at #iccv2021
Paper: https://t.co/jbScyR81KM
Project: https://t.co/s3fCOtuVil
Code: https://t.co/0hxeja8Vvy
@YVinker Nir Zabari @YHoshen