Scalable Set Encoding with Universal Mini-Batch Consistency and Unbiased Full Set Gradient Approximation

Official implementation of "Robust Molecular Property Prediction via Densifying Scarce Labeled Data" (GenBio Workship @ ICML 2025).

Training-free Post-training Efficient Sub-quadratic Complexity Attention. Implemented with OpenAI Triton.

LLM KV cache compression made easy

Vim-fork focused on extensibility and usability

OpenAI Triton Implementation of Streaming LLM

A Pytorch implementation of the optimal transport kernel embedding

[GCPR 2023 | CVPR 2023 Workshop] Self-Supervised Representation Learning on Point Clouds

Betty: an automatic differentiation library for generalized meta-learning and multilevel optimization

[ICML 2023] Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining

The PatchCamelyon (PCam) deep learning classification benchmark.

The world’s fastest framework for building websites.

[CVPR 2022] MPViT:Multi-Path Vision Transformer for Dense Prediction

Rank-1 update and downdate of Cholesky factorization

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (V…

Improving predictions of Bayesian neural nets via local linearization, AISTATS 2021

BackPACK - a backpropagation package built on top of PyTorch which efficiently computes quantities other than the gradient.

Collisions at high-energy particle colliders are a traditionally fruitful source of exotic particle discoveries. Finding these rare particles requires solving difficult signal-versus-background cla…

Pytorch implementation of "Large-Scale Meta-Learning with Continual Trajectory Shifting" (ICML 2021)

Code for experiments to learn uncertainty

[ACL 2021] Learning to Perturb Word Embeddings for Out-of-distribution QA

Official Code: Estimating Model Uncertainty of Neural Networks in Sparse Information Form, ICML2020.

Lightweight MCMC sampling for PyTorch Models aka My Corona Project

Implementations of many meta-learning algorithms to solve the few-shot learning problem in Pytorch

Database for AI. Store Vectors, Images, Texts, Videos, etc. Use with LLMs/LangChain. Store, query, version, & visualize any AI data. Stream data in real-time to PyTorch/TensorFlow. https://activelo…

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Wrecking sandwich traders for fun and profit