This is by no means a comprehensive list of AI research, tools, and products. However, it's a decent start to find what you may be looking for.
Stanford AI Index Report 2022
State of AI Report 2022
Succing Summary of Reinforcement Learning
AI Blog by Jürgen Schmidhuber
Annoted History of AI and Deep Learning
Midjourney: Images
Synthesia: Make AI Avatars
Descript: Make Voice-Overs on Text
Teleprompter: Listends to meetings and makes quote suggestions
Runway: All things Video / Content Editing
Vivid AI: AI Wingman to help build your web app
Twitter List
Train GPT on Arxiv Papers
Animate 2D Image
Perplexity: ChatGPT Competitor
Galileo AI: AI for UI Design
FileChat
ChatGPT Chrome Extensions
Copy.ai: Writing tool
Tome App: Creates powerpoints
Soundraw.io: Create music
Kaiber: Make videos
Fliki: Create Reels (videos)
Fireflies: Note taking
Gradio: ML model UI plug-in
Hebbia: Internal Business Search
Rose AI: ChatGPT for Data
Meta: CICERO
- The first AI to play at a human level in Diplomacy, a strategy game that requires building trust, negotiating and cooperating with multiple players.
- blog
Make-A-Video: Text-to-Video Generation without Text-Video Data
DALL-E: Zero-Shot Text-to-Image Generation
Pix2Pix Zero
DeepFaceLab: Deepfakes
DeepFaceLab Fork
DeepFaceLive: Deepfakes
Make-A-Video: Text-to-Video Generation without Text-Video Data
Galactica: A Large Language Model for Science
VALL-E: Generate voice snippets with only a few seconds of input audio
Whisper: OpenAI Speech Recognition
HuggingFace Stable Diffusion: Text2Img, Img2Img
BLIP-2: Image to Text
DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature
KnowledgeGPT: Question-answering tool for documents, includes citations in response
Fini: Turn your knowledge base into an AI Chat bot
BioGPT
BARD: ChatGPT Competitor
Runway's Diffusion-based video generation tool
Retreival-Enhanced Large Language Models
An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion
High Resolution Image Synthesis with Latent Diffusion Models
Multimodal Neurons in Artificial Neural Networks
Parsel: A Unified Natural Language Framework for Algorithmic Reasoning
Training Compute-Optimal Large Language Models
- Google investigates optimal training for language models, finds that models are undertrained (i.e. data is the bottleneck, not compute)
- paper
Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets
- OpenAI & Google: Improving to perfect generalization well past overfitting
- paper
Denoising Diffusion Probabilistic Models
Denoising Diffusion Implicit Models
Pseudo Numerical Methods for Diffusion Models on Manifolds
Fast Sampling of Diffusion Models with Exponential Integrator
Broken Neural Scaling Laws
- Modelling scaling behaviors of deep neural networks
- paper
Deepmind: Discovering novel algorithms with AlphaTensor
Deepmind: Accelerating fusion science through learned plasma control
PrefixRL: Optimization of Parallel Prefix Circuits using Deep Reinforcement Learning
Large Language Models can Self-Improve
Emergent Abilities of Large Language Models
Self-Instruct: Aligning Language Model with Self Generated Instructions
OpenAI: Learning to play Minecraft with video pre-training (VPT)
SparseGPT: GPT-family models can be pruned 50%+ sparsity in one-shot, without retraining and minimal loss of accuracy
OPT-IML: An instruction-tuned LLM for open-use
Dreambooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation
- mitigates the coherence problem in image generation (consistent generations) by binding a unique identifier to a fine-tuned model. This method powers the AI avatar creators, game asset generators, AI movies, and more.
- paper
Deep Learning and Computational Physics (Lecture Notes)
Diffusion Language Models
Does Compressing Activations Help Model Parallel Training?
Wolfram Alpha x ChatGPT
Data Distillation
Adjusting Biased Samples
- Easy to use framework for weighting data and evaluating its biases with and without adjustments
- github
- blog
Circumvent Context Size Limits
LLM Refinement Process, reduce mistakes
Multimodal Deep Learning
A Deep-Learning-Based Multi-Modal Sensor Fusion Approach for Detection of Equipment Faults
ChatGPT Political Biases
Pix2Pix
Deep Learning Tuning Playbook
3D Image Inpainting
Detecting Watermarks on GPT
Transformers: How "emergent abilities" are unlocked by scaling up language models
Google MusicLM
Prompting Strategy for High Performance
Dramatic NanoGPT Optimization
Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
Advanced Topics in MultiModal Machine Learning
Toolformer: An LM that teaches itself to use various tools in a self-supervised way
Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents
Categorical Archive of ChatGPT Failures
Optimizing Hard Text Prompts
Data Selection Framework for LMs
Rigorously Evaluating ChatGPT
Noise2Music: Generate 30 second music clips from text
Offsite Tuning: Transfer Learning without the Model
Mango: Hyperparameter Tuning
Graph Machine Learning