Éloi Zablocki
@EloiZablocki
Followers
568
Following
656
Media
59
Statuses
346
Research scientist @valeoai Alumnus @Sorbonne_Univ_ (team @mlia_isir) | @ENS_ParisSaclay (MVA) | @Polytechnique
Joined August 2012
Our recent research will be presented at #ICCV2025 @ICCVConference! We’ll present 5 papers about: 💡 self-supervised & representation learning 🌍 3D occupancy & multi-sensor perception 🧩 open-vocabulary segmentation 🧠 multimodal LLMs & explainability https://t.co/Tg0Vx3oS94
1
1
8
Really excited to talk about our recent open-source vision foundation model: Franca with @v_pariza. Thanks to the Cohere Vision community @cataluna84 and @Arkhymadhe for the invite. Join us on 23rd Sept at 5pm CET to gain insights on training large scale models.
Our Computer Vision Group is looking forward to hosting @shawshank_v and Valentinos Pariza for a presentation of "Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning." Thanks to @cataluna84 and @Arkhymadhe for organizing this guest speaker session!
1
6
21
Ep#12 with Florent Bartoccioni from @valeoai on VaViM and VaVAM: Autonomous Driving through Video Generative Modeling https://t.co/T5lmAioFI5 Co-hosted by @chris_j_paxton & @micoolcho
4
5
26
📢 We have a PR[AI]RIE PhD position opening @inria_paris co advised with R. de Charette & @tuan_hung_vu [please distribute] 💡Topic: Physics-Grounded Vision Foundation Models ⏳Application deadline: 20 May 2025 🗓️ Start date: Fall 2025 📝Detailed description: linked below
1
19
65
LLM-wrapper: Black-Box Semantic-Aware Adaptation of VLMs for Referring Expression Comprehension (REC), by @AmaiaCardiel, @EloiZablocki, @EliasRamzi, @oriane_simeoni, @quobbe Boost VLM perf by tuning an LLM to reason on its outputs! It's black-box🔒 & efficient⚡ (< 7h on 1 GPU)
1
1
2
π0 model for autonomous driving 🦾🚗 Nice job @valeoai!
Video Features + Flow Matching = Driving 🚙 VaVAM’s action module draws from π0: a flow matching approach for action prediction Kudos to the π0 team! (@kvablack @svlevine @DannyDriess @ihorbeaver @chelseabfinn @brian_ichter @hausman_k, & @RemiCadene @m_olbap @alibert_s) [6/10]
1
4
75
🚗 Ever wondered if an AI model could learn to drive just by watching YouTube? 🎥👀 We trained a 1.2B parameter model on 1,800+ hours of raw driving videos. No labels. No maps. Just pure observation. And it works! 🤯 🧵👇 [1/10]
6
50
178
After 6+ months in the making and burning over a year of GPU compute time, we're super excited to finally release the "Ultra-Scale Playbook" Check it out here: https://t.co/mnC0UzZYsJ A free, open-source, book to learn everything about 5D parallelism, ZeRO, fast CUDA kernels,
110
705
4K
“They said it could not be done”. We’re releasing Pleias 1.0, the first suite of models trained on open data (either permissibly licensed or uncopyrighted): Pleias-3b, Pleias-1b and Pleias-350m, all based on the two trillion tokens set from Common Corpus.
45
154
750
📢 Exciting opportunity alert! We ( https://t.co/HTAe8sVRlR) just posted our annual research internship openings in computer vision & ML. Check out the openings and the great achievements by our past interns here:
valeo.com
We are building an artificial intelligence research center for automotive applications based in Paris, since 2017.
🌟 Calling all MSc students passionate about computer vision and ML! We’re offering research internships about diffusion models, multi-modal transformers, continual learning, & more. 4 exciting openings await! 🔗 Learn more: https://t.co/HNVWpI5QJ3 RT to spread the word! 🙌
0
5
17
Self-supervised learning is fantastic for pretraining, but can we use it for other tasks (kNN classification, in-context learning) & modalities, w/o training & by simply using its gradients as features? Enter 🍄FUNGI - Features from UNsupervised GradIents #NeurIPS2024 🧵
9
49
334
🚨 Interested in domain adaptation and generalization for 3D data? 🚨 It’s tough to keep up with all the new amazing work. That’s why I’ve curated a comprehensive, easy-to-navigate list of publications. 📚 👉 Repo: https://t.co/k9a64VG3od
1
7
13
🍕 Heading to Milan for #ECCV2024! Our team has put together a mega-thread of our papers. I'll be presenting works on LLMs/VLMs, motion forecasting, and corner-case generation for autonomous driving. Looking forward to great discussions!
Preparing to be🚅@eccvconf ? Take a look at our recent works that we’ll present there. This year we're happy to share our results covering topics such as forecasting, tracking, domain adaptation, VLMs & LLMs and much more! Find out more 👇 and come meet us at #ECCV2024
0
2
12
🎬 video: https://t.co/0MPi66EbB0 📜 paper: https://t.co/EZAvkLpXgn 💻 website: https://t.co/djVAffAHLp ⌨️code:
github.com
A Unified Framework for scalable Vehicle Trajectory Prediction, ECCV 2024 - vita-epfl/UniTraj
1
0
2
UniTraj: A Unified Framework for Scalable Vehicle Trajectory Prediction With Lan Feng, Mohammadhossein Bahari, Kaouther Messaoud Ben Amor, Matthieu Cord (@quobbe), Alexandre Alahi (@AlexAlahi) Collaboration between @EPFL_en (VITA lab) and @valeoai Accepted at #ECCV2024
1
0
4
🤯 Also, we achieve SOTA results on nuScenes by simply training MTR on the whole UniTraj data.
1
0
0
📈 Regarding scaling, we find that performance gains remain in a linear regime with respect to dataset size. Even small models like Autobot (1.5M parameters) do not experience diminishing returns when trained on 2.3M trajectories.
1
0
0
📊 Dataset size is important, but we find other key factors using two diversity metrics: trajectory types and Kalman difficulties (discrepancy between future and simple linear extrapolation of the past). The Waymo dataset is more diverse 🌍
1
0
0