EvaByte: Efficient Byte-level Language Models at Scale

A toolkit implementing advanced methods to transfer models and model knowledge across tokenizers.

Squared Non-monotonic Probabilistic Circuits

A collection of useful scripts, templates, and examples for clusters using SLURM https://slurm.schedmd.com/

Solve puzzles. Improve your pytorch.

CausalFlows: A library for Causal Normalizing Flows in Pytorch

Efficient Transformers with Dynamic Token Pooling

Hydra is a framework for elegantly configuring complex applications

Command-line program to download videos from YouTube.com and other video sites

Program that constructs all oriented matroids of the given rank and on the given number of elements.

a python framework to build, learn and reason about probabilistic circuits and tensor networks

Code for UAI'19: Random Sum-Product Networks: A Simple and Effective Approach to Probabilistic Deep Learning

A Tool for Continuous Reachability Analysis.

Add icons to your Obsidian notes – Experimental Obsidian Plugin

Repository of PIXAR, a Pixel-based Auto-Regressive Language Model

Data and tools for generating and inspecting OLMo pre-training data.

List of papers on hallucination detection in LLMs.

How to Turn Your Knowledge Graph Embeddings into Generative Models

Code for paper "Automated mood disorder symptoms monitoring from multivariate time-series sensory data: Getting the full picture beyond a single number".

Semirings are a powerful unifying abstraction for dynamic programming algorithms

Sparse and structured neural attention mechanisms

arXiv LaTeX Cleaner: Easily clean the LaTeX code of your paper to submit to arXiv

The simplest, fastest repository for training/finetuning medium-sized GPTs.

Fast & Simple repository for pre-training and fine-tuning T5-style models

Inference code for Llama models

headquarters of the April team in Edinburgh

Obsidian plugin to automatically create a note from arXiv.org, acl anthology and semantic scholar.

CUDA kernels for generalized matrix-multiplication in PyTorch