Tom Sander (Ph.D.)

99 posts

Tom Sander (Ph.D.) banner
Tom Sander (Ph.D.)

Tom Sander (Ph.D.)

@RednasTom

AI Research Scientist at Meta Superintelligence Labs (FAIR). prev: Ph.D. @ FAIR & Polytechnique, engineering @ Polytechnique (X18), MVA @ ENS.

Paris, France Katılım Şubat 2022
191 Takip Edilen300 Takipçiler
Tom Sander (Ph.D.) retweetledi
Paul-Ambroise Duquenne
Paul-Ambroise Duquenne@duquenne_pa·
A couple of months after OmniASR, we’re excited to release OmniSONAR alongside OmniMT. OmniSONAR brings new training recipes for cross-lingual and cross-modal sentence encoders, enabling massively multilingual embeddings for text and speech. tinyurl.com/omnisonar 🧵 1/3
English
1
6
12
355
Tom Sander (Ph.D.)
Tom Sander (Ph.D.)@RednasTom·
Why does this matter? "Watermark Radioactivity." If we watermark specific documents post-hoc, we can detect if they are used to train future models or retrieved in RAG systems. It turns passive data into active tracers.
English
1
0
1
18
Tom Sander (Ph.D.) retweetledi
Pierre Fernandez
Pierre Fernandez@pierrefdz·
We're thrilled to share the open source release of Meta Seal, a comprehensive, SOTA, and MIT-licensed suite of AI watermarking research, models, & training code. Learn more in the 🧵 below and explore the artifacts here: facebookresearch.github.io/meta-seal
English
13
29
106
28.5K
Tom Sander (Ph.D.) retweetledi
Nikola Jovanović
Nikola Jovanović@ni_jovanovic·
There's a lot of work now on LLM watermarking. But can we extend this to transformers trained for autoregressive image generation? Yes, but it's not straightforward 🧵(1/10)
GIF
English
5
52
316
48.5K
Tom Sander (Ph.D.) retweetledi
Matteo Pirotta
Matteo Pirotta@teopir·
Exciting PhD position open at FAIR in Paris. We are looking for a candidate to join our team and contribute to advancing the field of AI, especially reinforcement learning. Find more details and apply below. Feel free to reach out to me by email. metacareers.com/jobs/192266079…
English
2
40
215
23.7K
Tom Sander (Ph.D.) retweetledi
Sagar Vaze
Sagar Vaze@Sagar_Vaze·
Our first LLM coding agent: Devstral. It's the SoTA OSS model on SWE-Bench Verified by a large margin, while being only 24B. It's still a research preview specific to SWE tasks. For regular code completion Codestral might still be the best bet. Let us know your feedback!
Mistral AI@MistralAI

Meet Devstral, our SOTA open model designed specifically for coding agents and developed with @allhands_ai mistral.ai/news/devstral

English
0
5
18
1.4K
Théo Uscidda
Théo Uscidda@theo_uscidda·
@NeurIPSConf overleaf has been down for several hours—maybe it's a sign that the main paper and supplemental deadlines should be merged into a single May 22 deadline, aligning it nicely with the position paper track deadline?
English
1
0
19
3K
NeurIPS Conference
NeurIPS Conference@NeurIPSConf·
📢 A reminder that the NeurIPS deadline for full paper submission is May 15th (Anywhere on Earth, AOE). We look forward to receiving your work, good luck to all submitting authors!
English
13
11
235
96.1K
Tom Sander (Ph.D.) retweetledi
Delong Chen (陈德龙)
Delong Chen (陈德龙)@Delong0_0·
This is my first paper done at FAIR. We show that adaptive visual token segmentation, especially in subobject-level (i.e., subwords in images), enables VLMs to have a better and faster learning of image understanding! arxiv.org/pdf/2402.14327
Delong Chen (陈德龙) tweet media
Paris, France 🇫🇷 English
1
12
59
5.2K
Tom Sander (Ph.D.)
Tom Sander (Ph.D.)@RednasTom·
I am at ICLR this week, presenting Watermark Anything on Saturday, and another work on radioactive benchmarks at the WM workshop on Sunday. Reach out if you want to discuss Watermarking, Distillation Protecion, Memorization, or else :)
Tom Sander (Ph.D.)@RednasTom

🎉Exciting news from @AIatMeta FAIR! We've released a Watermark Anything Model under the MIT license! It was announced yesterday: ai.meta.com/blog/meta-fair… Great project with @pierrefdz et al. ! We're close to hitting 1,000 stars on GitHub. Give it a try: github.com/facebookresear… 🚀

English
0
2
31
3.5K
Tom Sander (Ph.D.) retweetledi
Matthieu Meeus
Matthieu Meeus@matthieu_meeus·
Makes you wonder what kind of evidence they have, and got me thinking about this Neurips paper: arxiv.org/pdf/2402.14904. Any chance it's radioactive watermarking @RednasTom @pierrefdz?
Financial Times@FT

Distillation is a common practice in the AI industry but the concern was that DeepSeek may be doing it to build its own rival model, which is a breach of OpenAI’s terms of service. Here’s why: ft.trib.al/NydPc2b

English
0
1
9
490

Keşfet

@pierrefdz @theo_uscidda @NeurIPSConf @Meta @ENS_ULM @iclr_conf @GiulioBiroli @leventsagun