Oussama Zekri
@oussamazekri_
Followers
412
Following
2K
Media
22
Statuses
207
@ENS_ParisSaclay Maths department and @UW Research Intern | past : MVA @imperialcollege @Huaweifr @KyotoU_News | Research blog https://t.co/i7kGB3ffwv
Paris, France
Joined October 2023
🚀 Policy gradient methods like DeepSeek’s GRPO are great for finetuning LLMs via RLHF. But what happens when we swap autoregressive generation for discrete diffusion, a rising architecture promising faster & more controllable LLMs?. Introducing Score Entropy Policy Optimization
1
9
30
RT @AmbroiseOdonnat: 🚀To know more about LLM as Markov Chains, join in on June 19th at 6 pm CET (Paris time)!!😀 . Huge thanks to @itsmaddox….
0
3
0
RT @AmbroiseOdonnat: 💎It also works for the newest -- strongest Gemma3 models (👏🏽@ramealexandre @mblondel_ml)!
0
1
0
RT @Dorialexander: Ever more comforted in my assumption that Markov (1913) is the first language model.
0
27
0
RT @abenechehab: Excited to be heading to Singapore for #ICLR2025 this week! 🇸🇬 . I will be presenting our two latest works across the m….
0
4
0
Nice!.
TL;DR: we are excited to release a powerful new open-weight language model with reasoning in the coming months, and we want to talk to devs about how to make it maximally useful: we are excited to make this a very, very good model!. __. we are planning to.
0
0
1
RT @AmbroiseOdonnat: 🤗Thanks a lot, @haeggee and Prof. Martin Jaggi, for having me in the MLO group @EPFL this week to present "Large Langu….
0
2
0
RT @IevgenRedko: Our team open-sourced MANTIS: a foundation model for time series classification. It is lightweight, more efficient than c….
0
5
0
Every conference should do the same imo!.
Following a thorough investigation, the Program Chairs (PCs) decided to desk-reject 19 papers authored by confirmed highly irresponsible reviewers, which would have been accepted otherwise, in accordance with the previously communicated CVPR 2025 policies. 2/2.
0
0
2
RT @InceptionAILabs: We are excited to introduce Mercury, the first commercial-grade diffusion large language model (dLLM)! dLLMs push the….
0
990
0
RT @_GPaolo: Living organisms organize as loosely coupled hierarchies—while our AI systems remain rigid monoliths. What if we built AI th….
0
2
0
RT @Ji_Ha_Kim: I wrote a blog post for an introduction to stochastic calculus! I share my perspective and intuition behind Brownian motion,….
0
148
0
Nice ! And now you can do RLHF on this with this paper haha
Large Language Diffusion Models. Introduces LLaDA-8B, a large language diffusion model that pretrained on 2.3 trillion tokens using 0.13 million H800 GPU hours, followed by SFT on 4.5 million pairs. LLaDA 8B surpasses Llama-2 7B on nearly all 15 standard zero/few-shot learning
0
0
3
RT @qberthet: 🚨 New paper on regression and classification!. Adding to the discussion on using least-squares or cross-entropy, regression o….
0
60
0
RT @geoffnegiar: We just released our new website! . Our goal for now is to provide the easiest, fastest benchmarking tools for forecasting….
0
4
0