@TCebere

PhD 🧑🏻‍🎓 in Differential Privacy & Learning @Inria 🇫🇷 Alumnus @ENSdeLyon 🇫🇷, @upb1818 🇷🇴

tudorcebere.github.io
Joined July 2020
  • Tweets141
  • Following1.1K
  • Followers245
Nikhil Bansal and I wrote a book-length survey of algorithmic and geometric methods in discrepancy theory, and a draft of it is online arxiv.org/abs/2608.00140. This has been years in the making, and I’m excited to share it. More 👇

Discrepancy Theory: An Algorithmic and Geometric Perspective

Combinatorial discrepancy theory is a subject with roots in combinatorics, geometry, and number theory, and with numerous applications to mathematics and computer science. At its core, discrepancy...

arxiv.org
6
32
1
125
14,837
Tudor Cebere retweeted
Alex Meterez@alexmeterezJul 29
1/8 How well does the simplest model in optimization theory, the quadratic, actually describe LLM pretraining? Surprisingly well. We linearize a 150M model at checkpoints throughout training and the local Taylor expansion tracks the true loss for up to 10% of the training budget.
7
73
13
680
116,069
Our @NeurIPSConf #NeurIPS submission still has *no* meta-review. How cooked are we? Seriously: Can the hivemind help and let us know that someone is looking into this (and that maybe we are not the only ones)? [Please amplify for reach.]
6
7
59
8,295
Tudor Cebere retweeted
Mohammad Yaghini@MYaghiniApr 25
Can ML models be trained under fairness constraints with formal DP guarantees, without sacrificing utility? New paper at ICLR 2026: RaCO-DP Poster: Pavilion 4 (later today!) 🧵 (w/ @tudorcebere, Michael Menart, Aurélien Bellet, @NicolasPapernot )
1
3
1
14
1,961
AI agents face a critical security threat: Indirect Prompt Injection (IPI) attacks hijack agent behavior to steal credentials and cause financial loss. We built the first system-level defense for Computer-Use Agents that actually maintains utility! That is, don't change the model, change the system around it to guarantee control flow integrity! arxiv.org/pdf/2601.09923
3
19
3
110
10,397
Join Inria PreMeDICal in Montpellier for a 6-month internship working on the @opendp_org Library! 🇫🇷 💻 Combine theory + code 📈 Path to PhD/Eng roles 📩 Send your CV + start date More Details: researchers.lille.inria.fr/a…
1
4
187
Grateful to be named a 2025 Google PhD Fellow in privacy, safety, and security! I am excited to continue my research in differential privacy, auditing, and trustworthy AI.🎉🎉
Google.org
@Googleorg23 Oct 2025
🎉 We're excited to announce the 2025 Google PhD Fellows! @GoogleOrg is providing over $10 million to support 255 PhD students across 35 countries, fostering the next generation of research talent to strengthen the global scientific landscape. Read more: goo.gle/43wJWw8
A person with long hair is seen from behind working in a laboratory

ALT A person with long hair is seen from behind working in a laboratory

2
6
248
Tudor Cebere retweeted
Tiffany Ding@tifding10 Jul 2025
New paper: Conformal prediction for long-tailed classification🐒 arxiv.org/abs/2507.06867 🧑‍🌾 (plant enthusiast): Help me identify plants! 🤖 (existing conformal algs): Do you want sets that never include rare plants or sets that contain 100s of labels? 🧑‍🌾: Uhh… neither? A🧵(1/n)

Conformal Prediction for Long-Tailed Classification

Many real-world classification problems, such as plant identification, have extremely long-tailed class distributions. In order for prediction sets to be useful in such settings, they should (i)...

arxiv.org
2
8
2
48
11,406
ICML's election for their board of directors has begun. I've thrown my hat in the ring. Please consider voting for Gautam Kamath. I have experience with the governance of TMLR, COLT, and ALT, and I think I've demonstrated myself as a consciencious and engaged community member.
3
23
7
247
42,679
Learned a lot by working through this! Check out Sham's tweet summarizing our work and the full note
Sham Kakade
@ShamKakade621 Jun 2025
1/6 Infinite-dim SGD in linear regression is the strawman model for studying scaling laws, critical batch sizes, and LR schedules. We revisit (and simplify) its analysis using just linear algebra, making it easier to derive and reason about. No PSD operators. No tensor calculus.
2
14
1,302
Tudor Cebere retweeted
Roy Rinberg@RoyRinberg9 Jun 2025
RAGs are extremely useful, and yet there isn't an opensource RAG system for wikipedia (or I couldn't find it). So I built WikiRAG, a simple open-source github + hugging repo✨ Spin up your own RAG server for wikipedia in a single line. 🚗💨
4
2
1
15
1,195
Without its international students, Harvard is not Harvard. hrvd.me/IntStudents25t

Supporting Our International Students and Scholars

We have just filed a complaint and a motion for a temporary restraining order will follow. As we pursue legal remedies, we will do everything in our power to support our students and scholars.

harvard.edu
20,045
15,278
3,028
89,238
11,565,480
My team is hiring a Research Engineer!! If you would like to work with me on agentic privacy, apply here in the next two weeks job-boards.greenhouse.io/dee…
4
27
232
22,903
Tudor Cebere retweeted
OpenDP@opendp_org8 May 2025
SAVE THE DATE for the 2025 OpenDP Community Meeting on September 19th in Dublin, Ireland! We are thrilled to co-locate with the Eyes-Off Data Summit (@Oblivious) and can't wait to share more information on the agenda in the coming weeks - stay tuned 👀 sites.harvard.edu/opendp/eve…

2025 OpenDP Community Meeting

Join us on Friday, September 19th in Dublin, Ireland for the OpenDP Community Meeting!...

opendp.org
1
4
140
Excited to be in Singapore for ICLR, presenting our work on privacy auditing (w/ Aurélien & @NicolasPapernot). If you are interested in differential privacy/privacy auditing/security for ML, drop by (#497 26 Apr 10-12:30 pm) or let's grab a coffee! ☕ openreview.net/forum?id=xzKF…
2
25
1,867
This is an interesting summary of the OpenDP workshop "Differential Privacy Beyond Algorithms"!
OpenDP@opendp_org24 Mar 2025
📢 New Blog Alert! Read a recap of the DP Beyond Algorithms Workshop and learn more about how to be part of the community working on sociotechnical aspects of deploying DP! opendp.org/blog/reflections-…
4
165
🚨 We’ve trained on the entire web—now what? 🚨 Synthetic data holds promise as the next frontier but with caveats. Join us at the ICLR'25 workshop, "Will Synthetic Data Finally Solve the Data Access Problem?" this April in Singapore to discuss challenges and opportunities!
1
7
3
35
15,511
Spending the summer with the OpenDP team was an incredible experience—I highly recommend it! Feel free to reach out if you have any questions!
OpenDP@opendp_org16 Dec 2024
Applications for the 2025 OpenDP Interns and Visiting Fellows programs are now open! Deadline is February 1st 2025 for both programs - please share with your networks and colleagues! opendp.org/blog/applications…
6
423
Tudor Cebere retweeted
OpenDP@opendp_org13 Dec 2024
Introducing DP Wizard - a just released interactive tool to help explore different parameter choices and implement them with the OpenDP Library!  Read more about the tool in the blog below and please send us your feedback! opendp.org/blog/dp-wizard-ea…
3
7
193
Tudor Cebere retweeted
Neil Renic@NC_Renic9 Dec 2024
if academics are so smart why can't they connect to eduroam?
40
386
86
3,900
187,528
Load more