Niv Cohen

Post-doc at New York University

[Rohit Gandikota @rohitgandikota]

We tested several unlearning methods and found none of them really erase knowledge from the model - they simply hide it! 🧐

What does this mean? We must tread carefully with unlearning research within diffusion models🚨

Here is what we learned 🧵👇(led by @kevinlu4588)

[Niv Cohen @CohNiv]

Led by the amazing @kevinlu4588 !

[Kevin Lu @kevinlu4588]

Excited to share our paper “When Are Concepts Erased from Diffusion Models?” at @NeurIPSConf!

We introduce two conceptual models for erasure mechanisms in diffusion models, and a suite of probes to recover supposedly forgotten concepts.

Project website: https://t.co/pKQmjEASHK

[Thao Nguyen @thao_nguyen26]

If you are attending #ICML2025, check out our DataWorld workshop on Sat July 19. We have updated the website with more info on speakers & accepted papers! https://t.co/K3U540rqoe

Also happy to chat offline about all things ✨ data ✨

[Ben Recht @beenwrekt]

The NeurIPS paper checklist corroborates the bureaucratic theory of statistics. https://t.co/HHNhMSFzpv

[Niv Cohen @CohNiv]

🎤 Giving a keynote talk - "Anomaly Detection Beyond Pretrained Representations" tomorrow at 10:30am at #CVPR, VAND 3.0 workshop (Visual Anomaly and Novelty Detection). Come say hi! 👋

[Kevin Lu @kevinlu4588]

When we "erase" a concept from a diffusion model, is that knowledge truly gone? 🤔

We investigated, and the answer is often 'no'!

Using simple probing techniques, the knowledge traces of the erased concept can be easily resurfaced 🔍

Here is what we learned 🧵👇

[Neha Hulkund @NHulkund]

📣We are extending our deadline to May 31st!📣

Looking forward to seeing everyone's submissions :)

[Thao Nguyen @thao_nguyen26]

📢 Announcing our data-centric workshop at ICML 2025 on unifying data curation frameworks across domains!

📅 Deadline: May 24, AoE
🔗 Website: https://t.co/K3U540rqoe

We have an amazing lineup of speakers + panelists from various institutions and application areas.

[Anubhav Jain @anubhavj480]

Think your latent-noise diffusion watermarking method is robust? Think again!

We show that they are susceptible to adversarial attacks that only require one watermarked example and an off-the-shelf encoder. This attack can forge and remove the watermark with very high accuracy

[Omri Avrahami @OmriAvr]

I'm excited about the recent feature we've been working on at @pika_labs - action editing! 🏃 📹

Video editing methods often focus on spatial edits. Our method can directly edit the *action* of objects, even when the video already contains a distinctive motion (e.g, 👋).

[Niv Cohen @CohNiv]

In summary, WIND delivers robust, imperceptible watermarking for both AI-generated and natural images 🛡️🔏

For detailed results and insights, see our paper or code:
https://t.co/C73PCIWHCw

Led by Kasra Arabi in collaboration with @FeuerBenjamin @TealWitter @chegday

(5/5)

[Niv Cohen @CohNiv]

In our #ICLR2025 paper, we introduce WIND 🌬️
A method that embeds a distortion watermark directly in the diffusion noise! Our method ensures that the watermark in one image does not reveal information about the watermark in other images 🤫
📝 https://t.co/jmEw4g3scN

(1/5)

[Niv Cohen @CohNiv]

And there's more! By leveraging diffusion inpainting, WIND can also protect non-AI-generated images, opening exciting new pathways for copyright protection and authenticity verification across all digital content 🖼️🔑

(4/5)

[David Bau @davidbau]

Today we launch a new open research community with @wattenberg @viegasf @jacobandreas Ellie Pavlick @Brown_NLP and many others

It is called ARBOR.

please join us. 🧵

[Center for AI Safety @CAIS]

Existing AI benchmarks are becoming increasingly saturated.

Partnering with @scale_AI, we created Humanity's Last Exam, a new benchmark that tests the capabilities of models at the frontier of human knowledge and reasoning.

The results are out today:
https://t.co/JPC6dkeBKR

[Niv Cohen @CohNiv]

It was a pleasure to co-organize this workshop with the amazing @AvitalShafran! Can’t wait to see what she accomplishes next! 👏

[Avital Shafran @AvitalShafran]

And that's a wrap! Thank you for everyone that attended our @NeurIPSConf workshop on Red Teaming GenAI ☃️

1/5 🧵

[Niv Cohen @CohNiv]

@miniapeur Well, he most probably couldn't have taken a shower before physics

[Dan Hendrycks @hendrycks]

Have a question that is challenging for humans and AI?

We (@cais + @scale_AI) are launching Humanity's Last Exam, a massive collaboration to create the world's toughest AI benchmark.
Submit a hard question and become a co-author.
Best questions get part of $500,000 in prizes!

Deadline: Nov 1, 2024
Details: https://t.co/nsOEdIy3L1
Submit here: https://t.co/8lYmlFFUTp