Stars
AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.
[NeurIPS 2020] "The Lottery Ticket Hypothesis for Pre-trained BERT Networks", Tianlong Chen, Jonathan Frankle, Shiyu Chang, Sijia Liu, Yang Zhang, Zhangyang Wang, Michael Carbin
A modular RL library to fine-tune language models to human preferences
Instruction Following Agents with Multimodal Transforemrs
code release of research paper "Exploring Long-Sequence Masked Autoencoders"
Pre-Trained Language Models for Interactive Decision-Making [NeurIPS 2022]
Robust Speech Recognition via Large-Scale Weak Supervision
Model parallel transformers in JAX and Haiku
Large-scale pretrained models for goal-directed dialog
A latent text-to-image diffusion model
robomimic: A Modular Framework for Robot Learning from Demonstration
pytorch-implementation of Dreamer (Model-based Image RL Algorithm)
A beautiful, simple, clean, and responsive Jekyll theme for academics
Open source code for paper "On the Learning and Learnability of Quasimetrics".
Open source code for paper "Denoised MDPs: Learning World Models Better Than the World Itself"
Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)
Adaptive FNO transformer - official Pytorch implementation
Official implementation of "Multi-Task Learning as a Bargaining Game" [ICML 2022]
Continual Learning for Task-Oriented Dialogue Systems
An Empirical Investigation of Representation Learning for Imitation (EIRLI), NeurIPS'21
Implementation of Video Diffusion Models, Jonathan Ho's new paper extending DDPMs to Video Generation - in Pytorch
Official PyTorch implementation of GENESIS and GENESIS-V2