Max Conti @mlpc123
Research in NLP @illuintech, CS MSc @epfl Paris Joined September 2024-
Tweets36
-
Followers45
-
Following97
-
Likes126
I will be in Seoul with @marcantallard this week for ICML to present ModernVBERT 🇰🇷 Our poster session on Wednesday from 10:30PM in Hall A (#4104). Come chat with us!
Introducing ModernVBERT: a vision-language encoder that matches the performance of models 10× its size on visual document retrieval tasks! 👁️ Read more in the thread👇 (1/N)
if you write some really good code these days everyone thinks you did it with Claude Code. if you lose a lot of weight everyone thinks you did it with Ozempic. at least you can still be really good at dancing. they don't make ozempic or Claude Code for being good at dancing
okay maybe it's a good time? We have a small colbert model trained at pplx, it is a continue-training of pplx-embed-0.6b, so native multilingual, just made it open and added a section how to use MaxSim kernel: huggingface.co/perplexity-ai/…
Releasing my first kernel on @huggingface: MaxSim Late-interaction retrieval (ColBERT / PyLate) bottlenecks on materializing the full similarity matrix. This kernel avoids it by using tiled scoring with simdgroup_matrix (Metal) and WMMA. Result is 3–5× speedup compared to naive
Exciting publication alert!! Congrats to the whole team for the nice ideas and thorough execution 🎯 Hyped to take the time to deep-dive into every part :)
🚨 Do LLMs need to store everything they read in memory? To reduce KV cache size and improve decoding speeds, we propose Self-Pruned KV attention, a mechanism where the model learns to decide which KVs to write in the persistent KV cache, discarding all the rest! @AIatMeta🧵
We're releasing LateOn and DenseOn today. Two open retrieval models, 149M parameters each. LateOn (ColBERT, multi-vector): 57.22 NDCG@10 on BEIR. DenseOn (dense, single-vector): 56.20. Both beat models up to 4× larger We're open-sourcing the weights under Apache 2.0 🧵👇
The new generation of open state-of-the-art single and multi-vector retrieval models is here It's time, DenseOn with the LateOn 🎶 @LightOnIO releases models that leap past existing ones, and everything you need to do the same!
Update on the Late-Interaction Workshop @ ECIR2026: it was amazing! 🇳🇱 So cool to meet late-interactors IRL for the first time, I enjoyed every bit of that day 🤗 And congrats & huge thanks to the organizers again. Link to our work + poster below ⬇️
Looking forward to meeting fellow late interactors! 🤓🇳🇱
@mlpc123, @MaceQuent1 and I will be at the Late Interaction Workshop at #ECIR2026 on Thursday 🤗 See you in Delft 🇳🇱
Great work from @weaviate_io that compares the performances of text retrievers and multimodal ones. It appears that their errors are complementary, which makes their combination in hybrid search promising. Check their paper! arxiv.org/pdf/2602.17687
Most teams build RAG systems that only see text. But a lot of queries 𝘳𝘦𝘲𝘶𝘪𝘳𝘦 visual understanding to answer correctly. Most RAG systems treat PDFs the same way: OCR the text, chunk it, embed it, done. But that approach misses 𝗳𝗶𝗴𝘂𝗿𝗲𝘀, 𝘁𝗮𝗯𝗹𝗲𝘀, 𝘀𝗽𝗮𝘁𝗶𝗮𝗹
Love this from @weaviate_io!
Multimodal Hybrid Search is one line of code in Weaviate (in the code golf sense⛳️) Search with ColModernVBERT page image vectors, Arctic 2.0 text transcription vectors, and BM25! This is an especially interesting case of a multimodal, multimodel 😆 future of search. Derive
Most practicionners would agree that text embeddings should be "contextual" - ie. they should encode a passage w.r.t. the wider scope of the entire document the passage stems from; "They beat the British" could refer to football or french history without further context... In ConTEB (arxiv.org/abs/2505.24782), we highlight the standard failure modes of embedding models on retrieval tasks that require context to be properly embedded. We also propose a training strategy that extends standard "late chunking" to teach models to infuse embeddings with just the right amount of contextual knowledge to optimize retrieval. Super happy to see some new work by @perplexity_ai on contextual embedding models. They eval on ConTEB and use our in-sequence contrastive loss, along with a ton of cool techniques in multiple phases of training. Love the work @bo_wangbo and will read in details, but super happy to see one more stone towards contextual embedding models, in the path already traveled by @hxiao and @jxmnop ! Link to the paper: arxiv.org/abs/2602.11151…
Very proud of this paper, we lead many more experiments since the first release. I think we made a pretty complete analysis of what is currently possible with the benchmark ! Thanks again to everyone involved !
Two months ago, we released ViDoRe V3 benchmark. Now, the full paper is out! 📄arxiv.org/abs/2601.08620 Here is a recap of the benchmark along with a comprehensive breakdown of our findings on multimodal RAG. 🧵 (1/N)
In our EMNLP 2025 Oral paper with @mlpc123, we propose an extension to Late Chunking and demonstrate how we can embed contextual information within passage embeddings... and why it's often very useful to improve document retrieval! (9/15) arxiv.org/abs/2505.24782
And happy to see our dear ModernVBERT competing with models much larger on it!
So happy for you that this project finally sees the day, more than a simple extension of the previous versions! Congrats for the hard work @MaceQuent1 @antonio_loison 🥳
📢 ViDoRe V3, our new multimodal retrieval benchmark for enterprise use cases, is finally here! It focuses on real-world applied RAG scenarios using high-quality human-verified data. huggingface.co/blog/QuentinJG… 🧵(1/N)
@ManuelFaysse Friday 7th 10h45am! Looking forward :))
@JinaAI_ Why did you organize the BoF at the same time as a Information Extraction and Retrieval Oral session 😢 Will be presenting our work that builds on top of Late Chunking there instead :)
I'll be in Suzhou next week to present this project as an Oral at #EMNLP2025! 🥳 Let me know if you're there and wanna get in touch, or if you know anyone who'd be interested :) Looking forward! 🙌🇨🇳 x.com/mlpc123/status…
🕺Super happy to release our latest work with @ManuelFaysse: in our paper "Context Is Gold to Find the Gold Passage", we share all our findings on how to train embedding models to meaningfully include doc-wide context into chunks - leading to convincing results! 🧑🍳 🧵1/N
Looking forward to what @ManuelFaysse @pteiletche and @antonio_loison will be cooking next, and to the next one with @MaceQuent1 😉🤝
Besides our main results, it was also really interesting to look at some understudied training dynamics in more details Our findings suggest that we've been using visual encoders far below their potential, and I think we can expect a lot of improvements building on top of this!
Super excited to finally release this 🕺 What a fun project that was with these guys, s/o to them for all the great work!! x.com/pteiletche/sta…
Introducing ModernVBERT: a vision-language encoder that matches the performance of models 10× its size on visual document retrieval tasks! 👁️ Read more in the thread👇 (1/N)
Paulo Moura @paulomouraj
117 Followers 187 Following PhD Student on Information Retrieval @LightOnIO
Amélie Chatelain @AmelieTabatta
688 Followers 310 Following Head of Training & Inference @LightOnIO | PhD in physics of the universe 💫. Interested in IR, Evals & synthetic data. I also like gaming 🎮 and lifting 🏋️♀️.
search founder @n0riskn0r3ward
2K Followers 3K Following Search Relevance Engineer at @turbopuffer passionate about AI and search tech. Opinions my own.
Negar Foroutan @negarforoutan
762 Followers 944 Following Research Scientist, Google Research. Prev: #NLProc PhD student @EPFL_en, Multilingual NLP
Raphaël Sourty @raphaelsrty
1K Followers 926 Following AI @LightonIO Language Models, Information Retrieval, Knowledge Distillation PhD
Nishant Subramani @nsubramani23
922 Followers 2K Following Interpretability PhD student @LTIatCMU // steering vectors inventor // Prev: intern @msftresearch @google, predoc @allen_ai // @BVB supporter // vegan 🌱
Ben Clavié @bclavie
7K Followers 1K Following regressing linearly on a daily basis. wife guy who does retrieval. research @mixedbreadai, prev answerdotai - dms work randomly, responses might be very slow
Deepak Vijaykeerthy @deepakvijayke
349 Followers 2K Following Research Engineer: Post-training LLMs, Agents, & Evals. Building small training playgrounds: https://t.co/X09QlcjdTt Opinions are my own.
Aurélien-Morgan @AurelienMorgan_
100 Followers 3K Following Building `pip install retrain-pipelines` @ https://t.co/W3EMRzlG8u, ML-Eng-centric OSS DAG engine & WebConsole for neocloud jobs. Wandering around. Mind if I do.
🇵🇰 Abdul Ahad S... @aahad_main
77 Followers 2K Following لَا اِلہَ اِلَا اللہُ مُحَمّدُ رَسُولُ اللہ 不诱于誉, 不恐于诽 ,率道而行, 端然正己。
Mahmud ElHuseyni @mahmut_cv
42 Followers 3K Following AI Engineer @ https://t.co/e45EUHbhiC | MSc @medipolunv | RA @Tubitak | Vision + Language
Vincent Archer @ArcherVincent
427 Followers 2K Following Principal Data Scientist @ask_locala ⚡ #AdTech ⚙ #MachineLearning #BigData #Geomarketing #NLProc PhD in CS (Dr en informatique) https://t.co/jSlOWLx7jZ
Macé Quentin @MaceQuent1
45 Followers 55 Following Cifre PhD student @ ILLUIN Technology & MICS & LIG | Passionate about NLP research. Work on multimodal information retrieval & reliability of LLM based systems.
kind212 @kienduong212
3 Followers 509 Following
gooby_esq @gooby_esq
1K Followers 671 Following The real “Dan Price” | Director, Product Management @ Westlaw / Thomson Reuters | https://t.co/GMo86ocrru | Unofficial DSPy Hype Man |
fsommers @fsommers
106 Followers 553 Following
Antoine E. @antoine_edy
54 Followers 143 Following Research @illuintech | @CentraleLyon @UniOfSurrey
Jun Du @dujun001
114 Followers 4K Following
Joel Washington @joelwashington
1K Followers 6K Following Founder, Modern Talent (tech talent recruiting), Zero/Level (sold, ~$50m), Shift (went public, hit $1b rev & valuation). Fmr Pres., Belong SPAC (NASDAQ:BLNG)
dinos @din0s_
2K Followers 919 Following working on ai for scientific discovery. agent memory, information retrieval, evals, and lifting weights. incredibly optimistic about the future!
Hugues Sibille @sibille_hugues
234 Followers 170 Following PhD Candidate @ EPFL | 3D Computer Vision for Coral Reef Conservation. AI engineer passionated about LLMs, Agentic workflows and Information Retrieval
António Loison @antonio_loison
78 Followers 193 Following Researcher @hcompany_ai interested in multimodal RAG and GUI Agents.
xland1000 @xland1000
82 Followers 7K Following
Antoine Chaffin @antoine_chaffin
4K Followers 785 Following Solve search, solve everything @perplexity_ai
Manuel Faysse @ManuelFaysse
3K Followers 435 Following NLP Research @GoogleDeepmind Prev: @AIatMeta @epfl
Tony Wu @tonywu_71
2K Followers 550 Following Multimodal, RAG, Agents | ColPali co-first author | @centralesupelec 🇫🇷 x @Cambridge_Uni 🇬🇧 | Core Researcher at @hcompany_ai 🧑🏻💻
thijs @cdngdev
19K Followers 1K Following 20yo @stanford • robotics @openai 🤖 formerly @hackwithtrees @arena @discord @a16z
Julian Killingback @Julian_a42f9a
437 Followers 91 Following PhD Student at UMass Amherst working on Information Retrieval and Natural Language Processing
Noam Brown @polynoamial
175K Followers 958 Following Researching reasoning @OpenAI | Co-created Libratus/Pluribus superhuman poker AIs, CICERO Diplomacy AI, and OpenAI o-series 🍓 reasoning models
alex zhang @a1zhang
38K Followers 1K Following phd student @mit_csail @nlp_mit, previously undergrad @princeton 🫵🏻 go participate in the @GPU_MODE kernel competitions!
MrPhi @MonsieurPhi
39K Followers 375 Following Docteur en philosophie, ce qui ne soigne pas grand chose. Je fais des vidéos sur YouTube. Nanofictions : @Nano_Phi Discord : https://t.co/r5ME6x5LMo
Paulo Moura @paulomouraj
117 Followers 187 Following PhD Student on Information Retrieval @LightOnIO
Perplexity @perplexity_ai
506K Followers 66 Following Curiosity changes everything. Download our free app on iOS, Mac, Windows, and Android.
ParisMachineLearning @ParisMLgroup
5K Followers 838 Following We're the Paris ML Meetup. One of world's largest (8300+) https://t.co/z6d9xH9wTC @BardolFranck, @JackieForien, @CFalguiere & @IgorCarron
Linkup @Linkup_platform
1K Followers 705 Following We're building the next generation web search engine, connecting your applications and AI agents to the internet. https://t.co/a6nJdu6jfZ
Nicolas Boizard @N1colAIs
316 Followers 369 Following Final-year PhD student | EuroBERT, EuroLLM 🇪🇺 , BidirLM-Omnimodal 😎 Universal Logit Distillation Loss ⚗️ and more...
Amélie Chatelain @AmelieTabatta
688 Followers 310 Following Head of Training & Inference @LightOnIO | PhD in physics of the universe 💫. Interested in IR, Evals & synthetic data. I also like gaming 🎮 and lifting 🏋️♀️.
Pierre Colombo @PierreColombo6
537 Followers 1K Following Omni-modal AI researcher. Creator of ColPali, BidirLM-omi, EuroBERT & EuroLLM. Ex-Prof@ Centrale (Paris-Saclay) · Ex-CSO@ https://t.co/ncQ9gT2rpk (legaltech - SaulLM)
Negar Foroutan @negarforoutan
762 Followers 944 Following Research Scientist, Google Research. Prev: #NLProc PhD student @EPFL_en, Multilingual NLP
Antoine Bosselut @ABosselut
4K Followers 614 Following Helping machines make sense of the world. Asst Prof @ICepfl; Before: @stanfordnlp @allen_ai @uwnlp @MSFTResearch #NLProc #AI
Xiang Yue @xiangyue96
7K Followers 935 Following Superintelligence @Meta. Training & evaluating foundation models. Previously @LTIatCMU @osunlp. Opinions are my own.
Xueguang Ma @xueguang_ma
1K Followers 762 Following PhD student at @uwaterloo. Scaling search and reasoning for agents. Prev. intern at @Meta, @MSFTResearch, @amazon
Sebastian Raschka @rasbt
512K Followers 1K Following ML/AI research engineer. Ex stats professor. Author of "Build a Large Language Model From Scratch" (https://t.co/O8LAAMRzzW) & reasoning (https://t.co/5TueQKx2Fk)
Demis Hassabis @demishassabis
1.9M Followers 179 Following Nobel Laureate. Co-Founder & Chair @GoogleDeepMind; Chief Scientist of Alphabet. Founder & CEO @IsomorphicLabs. Building the future...
Yann LeCun @ylecun
1.3M Followers 797 Following Founder/Chair, AMI Labs; Professor, NYU; Partner, 224 Ventures; Ex-Chief AI Scientist, Meta. Researcher in AI, ML, Robotics, etc. ACM Turing Award Laureate.
Raphaël Sourty @raphaelsrty
1K Followers 926 Following AI @LightonIO Language Models, Information Retrieval, Knowledge Distillation PhD
François Fleuret @francoisfleuret
55K Followers 482 Following Research Scientist @meta (FAIR), Prof. @Unige_en, co-founder @neural_concept_. I like reality.
Nishant Subramani @nsubramani23
922 Followers 2K Following Interpretability PhD student @LTIatCMU // steering vectors inventor // Prev: intern @msftresearch @google, predoc @allen_ai // @BVB supporter // vegan 🌱
Cohere Labs @Cohere_Labs
27K Followers 269 Following @Cohere's research lab and open science initiative that seeks to solve complex machine learning problems. Join us in exploring the unknown, together.
Discrete Diffusion Re... @diffusion_llms
2K Followers 0 Following 📚 Journal club on discrete diffusion models 🎥 Replays available on YouTube! Contact: [email protected] Hosted by @ssahoo_, @jdeschena, @zhihanyang_
Keller Jordan @kellerjordan0
20K Followers 473 Following CIFAR-10 fanatic Pretraining @OpenAI OpCo LLC. Personal opinion poster
Niels Rogge @NielsRogge
24K Followers 745 Following ML Engineer @huggingface. Building https://t.co/z4nyO4pjVE. @KU_Leuven grad. General interest in machine & deep learning. Making AI more accessible for everyone!
elie @eliebakouch
25K Followers 5K Following research @PrimeIntellect (prev: @huggingface) anon feedback: https://t.co/JmMh7Sfvxd
Andi Marafioti @andimarafioti
9K Followers 710 Following leading multimodal research @huggingface (prev @unity)
Macé Quentin @MaceQuent1
45 Followers 55 Following Cifre PhD student @ ILLUIN Technology & MICS & LIG | Passionate about NLP research. Work on multimodal information retrieval & reliability of LLM based systems.
Illuin Technology @illuintech
159 Followers 15 Following Designing and building the future of tech solutions to strategic challenges. Fields : #AI, #NLP, #DataEngineering, #DevOps, #MobileDev, #AR, #VR, #Chatbots
Ai2 @allen_ai
86K Followers 448 Following Breakthrough AI to solve the world's biggest problems. › Join us: https://t.co/MjUpZpKPXJ › Newsletter: https://t.co/k9gGznstwj
Alexander Doria @Dorialexander
24K Followers 4K Following building open ai infrastructure @pleiasfr — χαλεπὰ τὰ καλά
tobi lutke @tobi
513K Followers 2K Following Shopify CEO by day, Dad in evening, hacker at night, Aspiring comprehensivist. + qmd !
Laude Institute @LaudeInstitute
4K Followers 426 Following Laude Institute backs computer science researchers turning research into real-world impact. // @LaudeVentures
Florian Brand @xeophon
18K Followers 843 Following evals @PrimeIntellect | open models @interconnectsai
AI at Meta @AIatMeta
855K Followers 367 Following Together with the AI community, we are pushing the boundaries of what’s possible through open science to create a more connected world.
Sara Hooker @sarahookr
65K Followers 11K Following Building intelligence that evolves @adaption_ai. Built @Cohere_Labs, @GoogleBrain, @GoogleDeepmind. ML Efficiency, Multimodal\lingual.
Julien Chaumond @julien_c
89K Followers 1K Following Co-founder and CTO at @huggingface 🤗. ML/AI for everyone, building products to propel communities fwd. @Stanford + @PolytechniqueYou might like
