@framart1

Assistant professor at École Polytechnique working on Trustworthy AI | he/him Other account: 🦋 mgubri

 Palaiseau, France
gubri.eu
Joined December 2012
  • Tweets785
  • Following811
  • Followers397
68Media
🦹💥 How to detect if my LLM was stolen or leaked? 🤖💥 I am delighted to announce TRAP 🪤, our new #ACL2024 findings paper ☝️ We showcase how to use adversarial prompt as model fingerprint for LLM. A thread 🧵 ⬇️⬇️⬇️
1
4
3
23
4,267
Interested in multilingualism and watermarking? Come now at poster 194! #EMNLP2026
Martin Gubri @EMNLP@framart1Mar 27
🌍We've made LLM watermarking equally robust across all languages we studied. Scaling to 100+ languages! Even sota watermarks can be removed by translating to another language, eg. Tamil. This hits hardest in low-resource languages, where moderation tools are already weak. 🧵
Translation attack against LLM watermarking.

ALT Translation attack against LLM watermarking.

Watermarking robustness across languages. Almost the same AUC x languages.

ALT Watermarking robustness across languages. Almost the same AUC x languages.

3
17
511
Martin Gubri @EMNLP retweeted
Asim Mohamed@amohamed264Aug 29
🎉 Proud to share our work on multilingual LLM watermarking got accepted to Findings of EMNLP 2026! More details in the post below. Preprint: arxiv.org/abs/2510.18019 (to be updated with the camera-ready version) #EMNLP2026 #LLMWatermarking #TextWatermarking #TrustworthyAI

Is Multilingual LLM Watermarking Truly Multilingual? Scaling...

Multilingual watermarking aims to make large language model (LLM) outputs traceable across languages, yet current methods still fall short. Despite claims of cross-lingual robustness, they are...

arxiv.org
Martin Gubri @EMNLP@framart1Mar 27
🌍We've made LLM watermarking equally robust across all languages we studied. Scaling to 100+ languages! Even sota watermarks can be removed by translating to another language, eg. Tamil. This hits hardest in low-resource languages, where moderation tools are already weak. 🧵
Translation attack against LLM watermarking.

ALT Translation attack against LLM watermarking.

Watermarking robustness across languages. Almost the same AUC x languages.

ALT Watermarking robustness across languages. Almost the same AUC x languages.

1
1
117
🎉 I am extremely excited to announce the biggest news of my career: I will join École Polytechnique as an assistant professor in September!
Polytechnique logo

ALT Polytechnique logo

LIX's offices

ALT LIX's offices

3
20
875
more replies
I am very grateful for the trust placed in me by Sonia Vanier and all the academics, and very excited to work with the current and future members of the ORAILIX lab.
1
1
53
A huge thanks to all my co-authors for their hard work, and to my co-workers, friends and family for their support! I look forward to the years ahead!!
3
53
Martin Gubri @EMNLP retweeted
Anmol Goel@anmgoelJun 29
I'll be at #ACL2026 in San Diego, presenting our recent works spanning AI safety! 📍 Poster (Jul 6) Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models arxiv.org/abs/2601.15220 (1/2)
4
2
1
20
1,513
Martin Gubri @EMNLP retweeted
Cornelius Emde@CorEmdeJul 5
Today at #ACL2026, we are presenting out MASEval library for multi-agent system evaluation. @anmgoel is in San Diego to present poster and live demo! 📍Grand Hall | Session 3: Oral/Posters/Demos B 🕑Sunday 2pm-3.30pm #MultiAgentSystem #AI #AIAgents #ACL #Evaluation
Cornelius Emde@CorEmdeMar 23
1/ Evaluating a single agent harness is hard. Evaluating a multi-agent system? That's a whole different problem. Most eval tools treat the model as the unit of analysis. But in multi-agent systems, the system is what matters. That's why we built MASEval 🧵 #Agents #AI #Eval
1
1
9
736
🏆🪩 DISCO just won the best paper award at ICLR'26-CAO! I am so proud of my co-authors! Huge congrats to @a_rubique, Benjamin, and @coallaoh!
Alexander Rubinstein@a_rubiqueMay 1
🎉 Our DISCO paper received a Best Paper Award at the ICLR 2026 “Catch, Adapt, and Operate” Workshop! Huge thanks to my coauthors Benjamin, @framart1, and @coallaoh, and to the CAO organizers, led by @sepidshs, and @RBCBorealis, for supporting the workshop!
2
10
372
🪩 If you are at #ICLR today, come the poster of our DISCO paper to discuss LLM efficient benchmarking with @a_rubique!
Alexander Rubinstein@a_rubiqueApr 24
Come check out our poster to learn a simple (but very effective) trick to speed up your LLM evaluation by 100×! ⏰ 15:15 to 17:45 today! 📍Pavilion 3 P3-#1015
8
203
1/ My contract at @parameterlab ended last week, after 2.5 years (since Sept 2023, with some collaboration before). I had the chance to lead research on trustworthy AI for LLMs alongside an incredible group of people (Neckarfront. All Tübingen researchers have to post it once!)
tubingen classical view

ALT tubingen classical view

2
15
1,160
more replies
5/ And to all my brilliant co-authors since 2023: @dnnslmr, @oodgnas, @hwaran_lee, Siwon, @HaritzPuerto, @tommasogreen, @anmgoel, @CorEmde, Ahmed, @esruzzetti, Salman, @amohamed264, @a_rubique, @adamdaviesnlp, @_elinguyen, @michael_simeone, Erik.
1
5
197
6/ Thank you for the hard work, the patience, and the brilliance! I learned so much from all of you. More to come soon :)
5
88
🎉Our privacy collapse paper has been accepted at #ACL 2026 (main)! Contextual privacy is fragile: fine-tune an LLM on benign data, and it can overshare personal information. This is silent: suites don't usually measure contextual privacy, which is a big issue with LLM agents.
Privacy collapse at ACL

ALT Privacy collapse at ACL

Anmol Goel@anmgoelFeb 3
🚨 Fine-tuning your model to be more helpful or empathetic might be making it less private, without you noticing. In our latest work, we show that benign fine-tuning can silently break contextual privacy in language models while safety & general capabilities appear intact. ⬇️
1
4
224
🌍We've made LLM watermarking equally robust across all languages we studied. Scaling to 100+ languages! Even sota watermarks can be removed by translating to another language, eg. Tamil. This hits hardest in low-resource languages, where moderation tools are already weak. 🧵
Translation attack against LLM watermarking.

ALT Translation attack against LLM watermarking.

Watermarking robustness across languages. Almost the same AUC x languages.

ALT Watermarking robustness across languages. Almost the same AUC x languages.

1
2
2
8
922
more replies
Kudos to Asim Mohamed for his first research paper! Paper: arxiv.org/abs/2510.18019 Code: github.com/asimzz/steam

Is Multilingual LLM Watermarking Truly Multilingual? Scaling...

Multilingual watermarking aims to make large language model (LLM) outputs traceable across languages, yet current methods still fall short. Despite claims of cross-lingual robustness, they are...

arxiv.org
2
98
Would love to hear thoughts from people working on watermarking :) cc @ni_jovanovic @sahar_abdelnabi @jonasgeiping
98
Load more