théo gigant

gigant

https://giganttheo.github.io/

AI & ML interests

multimodal

Recent Activity

authored a paper 2 days ago

Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation

upvoted a paper 3 days ago

Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation

submitted a paper 3 days ago

Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation

View all activity

Organizations

upvoted a paper 3 days ago

Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation

Paper • 2604.27263 • Published 11 days ago • 8

upvoted a paper 6 days ago

Targeted Neuron Modulation via Contrastive Pair Search

Paper • 2605.12290 • Published 13 days ago • 14

upvoted a paper 9 days ago

Long Context Pre-Training with Lighthouse Attention

Paper • 2605.06554 • Published 18 days ago • 29

upvoted a paper 11 days ago

Efficient Pre-Training with Token Superposition

Paper • 2605.06546 • Published 18 days ago • 44

upvoted a collection 9 months ago

Hermes 4 Collection

Collection

9 items • Updated Mar 2 • 104

upvoted a paper 9 months ago

Hermes 4 Technical Report

Paper • 2508.18255 • Published Aug 25, 2025 • 50

upvoted a paper 11 months ago

Should We Still Pretrain Encoders with Masked Language Modeling?

Paper • 2507.00994 • Published Jul 1, 2025 • 81

upvoted an article 11 months ago

Article

SmolLM3: smol, multilingual, long-context reasoner

eliebak, cmpatino, anton-l, edbeeching, m-ric, nouamanetazi, akseljoonas, guipenedo, hynky, clefourrier, SaylorTwift, kashif, qgallouedec, hlarcher, glutamatt, Xenova, reach-vb, ngxson, craffel, lewtun, loubnabnl, lvwerra, thomwolf

•

Jul 8, 2025

• 777

upvoted a collection 11 months ago

🧠 SmolLM3

Collection

Smol, multilingual, long-context reasoner • 14 items • Updated Oct 9, 2025 • 104

upvoted an article 11 months ago

Article

Efficient MultiModal Data Pipeline

ariG23498, lusxvr, andito, sergiopaniego, pcuenq

•

Jul 8, 2025

• 71

upvoted 2 articles about 1 year ago

Article

nanoVLM: The simplest repository to train your VLM in pure PyTorch

ariG23498, lusxvr, andito, sergiopaniego, merve, pcuenq, reach-vb

•

May 21, 2025

• 258

Article

Vision Language Models (Better, faster, stronger)

merve, sergiopaniego, ariG23498, pcuenq, andito

•

May 12, 2025

• 613

upvoted 2 papers about 1 year ago

Perception Encoder: The best visual embeddings are not at the output of the network

Paper • 2504.13181 • Published Apr 17, 2025 • 36

Summarization of Multimodal Presentations with Vision-Language Models: Study of the Effect of Modalities and Structure

Paper • 2504.10049 • Published Apr 14, 2025 • 2

upvoted 2 articles about 1 year ago

Article

Open R1: Update #3

open-r1

•

Mar 11, 2025

• 297

Article

Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM

ariG23498, merve, pcuenq, reach-vb

•

Mar 12, 2025

• 497

upvoted a paper about 1 year ago

EuroBERT: Scaling Multilingual Encoders for European Languages

Paper • 2503.05500 • Published Mar 7, 2025 • 81

upvoted 2 articles about 1 year ago

Article

Introducing EuroBERT: A High-Performance Multilingual Encoder Model

EuroBERT

•

Mar 10, 2025

• 147

Article

A Deepdive into Aya Vision: Advancing the Frontier of Multilingual Multimodality

saurabhdash, olivernan, ArashAhmadian, johndang-cohere

•

Mar 4, 2025

• 78

upvoted a paper over 1 year ago

Qwen2.5-VL Technical Report

Paper • 2502.13923 • Published Feb 19, 2025 • 218

théo gigant

AI & ML interests

Recent Activity

Organizations

gigant's activity

SmolLM3: smol, multilingual, long-context reasoner

Efficient MultiModal Data Pipeline

nanoVLM: The simplest repository to train your VLM in pure PyTorch

Vision Language Models (Better, faster, stronger)

Open R1: Update #3

Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM

Introducing EuroBERT: A High-Performance Multilingual Encoder Model

A Deepdive into Aya Vision: Advancing the Frontier of Multilingual Multimodality