Stars
8000Awesome papers on weight-space learning
Minimalistic 4D-parallelism distributed training framework for education purpose
Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.
Official code for NeurIPS 2024 paper "A Canonicalization Perspective on Invariant and Equivariant Learning".
Code for "The Empirical Impact of Neural Parameter Symmetries, or Lack Thereof". NeurIPS 2024 paper.
Automated Identification of Redundant Layer Blocks for Pruning in Large Language Models
Implementation of algorithms from the paper "Efficient Low Rank Gaussian Variational Inferencefor Neural Networks"
Pytorch implementations of Bayes By Backprop, MC Dropout, SGLD, the Local Reparametrization Trick, KF-Laplace, SG-HMC and more
Bayesianize: A Bayesian neural network wrapper in pytorch
Must-read Papers on Large Language Model (LLM) Continual Learning
Retrieval and Retrieval-augmented LLMs
Expressive Sign Equivariant Networks for Spectral Geometric Learning
The WeightWatcher tool for predicting the accuracy of Deep Neural Networks
Official implementation of the ICML 23 paper "Equivariant Polynomials for Graph Neural Networks".
Official implementation for Equivariant Architectures 978D for Learning in Deep Weight Spaces [ICML 2023]
This is an official implementation for "GRIT: Graph Inductive Biases in Transformers without Message Passing".
Domain-Specific Languages of Mathematics
Public source code for the paper "Analyzing Monotonic Linear Interpolation in Neural Network Loss Landscapes"
Model Zoos published at the NeurIPS 2022 Dataset & Benchmark track: "Model Zoos: A Dataset of Diverse Populations of Neural Network Models"
Recipe for a General, Powerful, Scalable Graph Transformer
A playbook for systematically maximizing the performance of deep learning models.
Instant neural graphics primitives: lightning fast NeRF and more
This is the official repo for the paper "Neural Set Function Extensions: Learning with Discrete Functions in High Dimensions", presented at NeurIPS 2022.
Code release for REPAIR: REnormalizing Permuted Activations for Interpolation Repair
[NeurIPS'22] Tokenized Graph Transformer (TokenGT), in PyTorch
Pen and paper exercises in machine learning