VILA - a multi-image visual language model with training, inference and evaluation recipe, deployable from cloud to edge (Jetson Orin and laptops)

HOPS - A user-friendly software for single-object photometry and exoplanet light-curve modelling

Fast Inference in Denoising Diffusion Models via MMD Finetuning

MuseTalk: Real-Time High Quality Lip Synchorization with Latent Space Inpainting

model adapter from avhubert to whisper

the dataset and code for "Flow-guided One-shot Talking Face Generation with a High-resolution Audio-visual Dataset"

Inspired by "Neural Networks Fail to Learn Periodic Functions and How to Fix It"

Effortless plugin and play Optimizer to cut model training costs by 50%. New optimizer that is 2x faster than Adam on LLMs.

Open Python 3 API for making your own Smash Bros: Melee AI that works with Slippi Online

Imitation learning from slippi replays.

Examples and guides for using the OpenAI API

Instruction Tuning with GPT-4

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

Foundation Architecture for (M)LLMs

Noah Research

Hash function quality and speed tests

Multi-Task Deep Neural Networks for Natural Language Understanding

Olympiad in Informatics solutions

Denoising Diffusion Implicit Models

Dual Diffusion Implicit Bridges for Image-to-Image Translation. ICLR 2023.

Google Brain AutoML

Simple and efficient RevNet-Library for PyTorch with XLA and DeepSpeed support and parameter offload

Image restoration with neural networks but without learning.

Convolutional Gaussian processes based on GPflow.

Implementation of 1D, 2D, and 3D FFT convolutions in PyTorch. Much faster than direct convolutions for large kernel sizes.

[ICLR'23] DiffuSeq: Sequence to Sequence Text Generation with Diffusion Models

Compressed numerical arrays that support high-speed random access