Ross Wightman's picture

Ross Wightman

rwightman

·

AI & ML interests

Computer vision, transfer learning, semi/self supervised learning, robotics.

Recent Activity

updated a model 1 day ago

timm/vit_so400m_patch16_siglip_gap_256.webli_i18n

updated a model 1 day ago

timm/vit_so400m_patch16_siglip_256.webli_i18n

updated a model 1 day ago

timm/vit_so400m_patch14_siglip_gap_384.webli

View all activity

Articles

Trick or ResNet Treat

Mamba Out

Tiny Test Models

Searching for better (Full) ImageNet ViT Baselines

MobileNet Baselines

MobileNet-V4 (now in timm)

Organizations

rwightman's activity

upvoted a paper 5 days ago

Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference

Paper • 2412.13663 • Published 7 days ago • 103

upvoted a paper 9 days ago

Phi-4 Technical Report

Paper • 2412.08905 • Published 13 days ago • 92

upvoted a paper 20 days ago

PaliGemma 2: A Family of Versatile VLMs for Transfer

Paper • 2412.03555 • Published 20 days ago • 118

upvoted a collection 20 days ago

Common Models

The first generation of models pretrained on Common Corpus. • 5 items • Updated 20 days ago • 27

upvoted 2 papers 21 days ago

MARS: Unleashing the Power of Variance Reduction for Training Large Models

Paper • 2411.10438 • Published Nov 15 • 13

Cautious Optimizers: Improving Training with One Line of Code

Paper • 2411.16085 • Published about 1 month ago • 15

upvoted an article about 1 month ago

Article

🤗 Serve any model with Inference Endpoints + Custom Handlers

By

•

Nov 22

• 3

upvoted 2 collections 3 months ago

RDNet

DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTs [ECCV 2024] • 9 items • Updated Oct 16 • 3

timm tiny test models

A collection of very small (~300-500k parameter) models at 160x160 resolution, for testing purposes. Trained on ImageNet-1k. • 13 items • Updated Oct 2 • 5

upvoted 2 articles 5 months ago

Article

MobileNet Baselines

By

•

Jul 26

• 23

Article

LAVE: Zero-shot VQA Evaluation on Docmatix with LLMs - Do We Still Need Fine-Tuning?

Jul 25

• 18

upvoted a collection 5 months ago

🍃 MINT-1T

Data for "MINT-1T: Scaling Open-Source Multimodal Data by 10x: A Multimodal Dataset with One Trillion Tokens" • 13 items • Updated Jul 24 • 56

upvoted 2 papers 6 months ago

PaliGemma: A versatile 3B VLM for transfer

Paper • 2407.07726 • Published Jul 10 • 68

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs

Paper • 2406.16860 • Published Jun 24 • 59

upvoted a collection 6 months ago

Cambrian Data

3 items • Updated Jun 25 • 10

upvoted a paper 6 months ago

An Image is Worth 32 Tokens for Reconstruction and Generation

Paper • 2406.07550 • Published Jun 11 • 55

upvoted a collection 6 months ago

MobileCLIP Models + DataCompDR Data

MobileCLIP: Mobile-friendly image-text models with SOTA zero-shot capabilities. DataCompDR: Improved datasets for training image-text SOTA models. • 22 items • Updated Oct 4 • 25

upvoted a collection 7 months ago

MobileNetV4 pretrained weights

Weights for MobileNet-V4 pretrained in timm • 17 items • Updated Sep 22 • 18

upvoted 2 papers 7 months ago

MobileNetV4 -- Universal Models for the Mobile Ecosystem

Paper • 2404.10518 • Published Apr 16 • 2

On the Efficiency of Convolutional Neural Networks

Paper • 2404.03617 • Published Apr 4 • 4