LAION-5B: An open large-scale dataset for training next generation image-text models
Explore this paper's citation graph
Summary
This work presents LAION-5B - a dataset consisting of 5.85 billion CLIP-filtered image-text pairs, of which 2.32B contain English language, and shows successful replication and fine-tuning of foundational models like CLIP, GLIDE and Stable Diffusion using the dataset, and discusses further experiments enabled with an openly available dataset of this scale.
- Type
- preprint
- Published
- 2022-10-16
- Cited by
- 5,499
- References
- 109
- Access
- Open access
- OpenAlex
- https://openalex.org/W4306820534
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:252917726
Keywords
Computer science, Robustness (evolution), Artificial intelligence, Image (mathematics), Scale (ratio)
References
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- SUN Database: Exploring a Large Collection of Scene Categories
- Describing Textures in the Wild
- Large scale image annotation: learning to rank with joint word-image embeddings
- ImageNet: A large-scale hierarchical image database
- ImageNet Large Scale Visual Recognition Challenge
- 3D Object Representations for Fine-Grained Categorization
- Learning Visual Representations using Images with Captions
- From image descriptions to visual denotations: New similarity metrics for semantic inference over event descriptions
- Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations
- Generative Adversarial Text to Image Synthesis
- Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs.
- Learning Visual N-Grams from Web Data
- StackGAN: Text to Photo-Realistic Image Synthesis with Stacked Generative Adversarial Networks
- The european language resources association
- Exploring the Limits of Weakly Supervised Pretraining
- Do Better ImageNet Models Transfer Better?
- Conceptual Captions: A Cleaned, Hypernymed, Image Alt-text Dataset For Automatic Image Captioning
- Model Cards for Model Reporting
- JUST: Large-Scale Multi-Tier Storage Infrastructure at the Jülich Supercomputing Centre
Cited by
- Retrieval-Augmented Diffusion Models
- Background Invariance Testing According to Semantic Proximity
- The Biased Artist: Exploiting Cultural Biases via Homoglyphs in Text-Guided Image Generation Models
- Deep Lake: A Lakehouse for Deep Learning
- Natural language supervision with a large and diverse dataset builds better models of human high-level visual cortex
- DreamFusion: Text-to-3D using 2D Diffusion
- Visualize Before You Write: Imagination-Guided Open-Ended Text Generation
- Adapting Pretrained Vision-Language Foundational Models to Medical Imaging Domains
- Unifying Diffusion Models' Latent Space, with Applications to CycleDiffusion and Guidance
- 1st Place Solution in Google Universal Images Embedding
- 6th Place Solution to Google Universal Image Embedding
- 5th Place Solution to Kaggle Google Universal Image Embedding Competition
- Language Does More Than Describe: On The Lack Of Figurative Speech in Text-To-Image Models
- Conditional Diffusion with Less Explicit Guidance via Model Predictive Control
- DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models
- Rickrolling the Artist: Injecting Invisible Backdoors into Text-Guided Image Generation Models
- AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities
- Fast Text-Conditional Discrete Denoising on Vector-Quantized Latent Spaces
- DreamArtist: Controllable One-Shot Text-to-Image Generation via Positive-Negative Adapter
- RoentGen: Vision-Language Foundation Model for Chest X-ray Generation
Related papers
- Researching robustness of information system for measuring of microcontrollers average power consumption
- Development and Design Optimization of 2Y Hexarotor with Robustness against Rotor Failure
- Robustness evaluation in analytical methods optimized using experimental designs
- Robust resource loading for engineer-to-order manufacturing
- Robust signal detection with nonstandard decision regions
- Robustness measurement for scalable switch fabric
- Proving programs robust
- A robustness enhancer for model-based controllers
- Robustness assessment of complex networks using the idle network
- Violations of robustness trade-offs