dsevero
Follow
Stars
Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking
Code and Examples for Relevant Search
Code for the paper 'Learning TSP Requires Rethinking Generalization' (CP 2021)
Official repository for "DiffAssemble: A Unified Graph-Diffusion Model for 2D and 3D Reassembly" accepted at CVPR2024
Code for "Positional Diffusion: Ordering Unordered Sets with Diffusion Probabilistic Models"
A collection of PDDL generators, some of which have been used to generate benchmarks for the International Planning Competition (IPC).
The Fast Downward domain-independent classical planning system
A fast Sudoku solver and generator with a benchmark suite for comparing the fastest known solvers.
Neural Networks and the Chomsky Hierarchy
Official PyTorch implementation for "Large Language Diffusion Models"
Exca - Execution and caching tool for python
Official Jax Implementation of MD4 Masked Diffusion Models
Code for the paper https://arxiv.org/abs/2402.04997
Implementation of the paper "Lossless Compression of Vector IDs for Approximate Nearest Neighbor Search" by Severo et al.
SymmetricDiffusers: Learning Discrete Diffusion on Finite Symmetric Groups
[ICLR 2025] Code for the paper "Beyond Autoregression: Discrete Diffusion for Complex Reasoning and Planning"
Helpful tools and examples for working with flex-attention
A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.
Lagrangian formulation of Doob's h-transform allowing for efficient rare event sampling
Maze navigation with MLM-U
Generate a random permutation with a given number of inversions
maze datasets for investigating OOD behavior of ML systems
Official implementation of NeurIPS'24 paper "Causal Language Modeling Can Elicit Search and Reasoning Capabilities on Logic Puzzles"