VeRA: Vector-based Random Matrix Adaptation

PaLI-3 Vision Language Models: Smaller, Faster, Stronger

HyperAttention: Long-context Attention in Near-Linear Time

Fast Feedforward Networks

Nougat: Neural Optical Understanding for Academic Documents

Retentive Network: A Successor to Transformer for Large Language Models

LLava: Visual Instruction Tuning

BloombergGPT: A Large Language Model for Finance

ImageBind: One Embedding Space To Bind Them All

Segment Anything

Are Emergent Abilities of Large Language Models a Mirage?

Synthetic Data Boosts ImageNet Classification

Unlimiformer: Long-Range Transformers with Unlimited Length Input

[Tutorial] Image Super Resolution without Photoshop

YOLO9000: Better, Faster, Stronger

NÜWA: Visual Synthesis Pre-training for Neural visUal World creAtion

Florence: A New Foundation Model for Computer Vision

DSSD: Deconvolutional Single Shot Detector

MAE: Masked Autoencoders Are Scalable Vision Learners

PVANet: Deep but Lightweight Neural Networks forReal-time Object Detection

SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers

R-FCN: Object Detection via Region-based Fully Convolutional Networks

Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

Pix2Seq: A Language Modeling Framework for Object Detection

Improved Regularization of Convolutional Neural Networks with Cutout

VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning

SSD: Single Shot MultiBox Detector

Barlow Twins: Self-Supervised Learning via Redundancy Reduction

MLP-Mixer: An all-MLP Architecture for Vision

YOLO: Unified, Real-Time Object Detection