There are 1 repository under compositionality topic.
Generative Adversarial Transformers
A framework for applied category theory in the Julia language
ACT community resources
[ECCV 2022] Compositional Generation using Diffusion Models
Experiments and data for the paper "When and why vision-language models behave like bags-of-words, and what to do about it?" Oral @ ICLR 2023
A curated list of Composable AI methods: Building AI system by composing modules.
[NeurIPS 2019] Code for the paper "Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularity"
Distributional Generalization in NLP. A roadmap.
[ICCV 2023] Unsupervised Compositional Concepts Discovery with Text-to-Image Generative Models
NeurIPS 2022 Paper "VLMbench: A Compositional Benchmark for Vision-and-Language Manipulation"
Tensor algebra abstract type interoperability setup
Official implementation and dataset for the NAACL 2024 paper "ComCLIP: Training-Free Compositional Image and Text Matching"
Multimodal Graph Network (MGN): Code repo, examples from the paper
A list of resources dedicated to compositionality
A Neural Compositional Paradigm for Image Captioning
Tool to measure tree-structuredness of the internal algorithm learnt by a transformer
Codebase for analysing compositional generalisation in NMT models, which allows you to run systematicity, productivity, substitutivity and idiom processing analyses.
Code accompanying the paper Measuring non-trivial compositionality in emergent communication
Code repository for Capacity, Bandwidth, and Compositionality in Emergent Language Learning (AAMAS 2020)
TensorFields with product topology using Grassmann element parameters
SVIB: Systematic Visual Imagination Benchmark
Living Survey for papers on Compositional Generalization in NLP
Code and results for experiments of CNNs (fairseq, Gehring et al., ICML, 2017) on SCAN dataset (Lake & Baroni, ICML, 2018)
Code for Recursive Neural Programs: A differentiable framework for learning compositional part-whole hierarchies and image grammars
Code acompanying the paper Developmentally motivated emergence of compositional communication via template transfer
Source code and supplementary material for DIRAC: Diffusion-Based Representation Learning for Modality-Agnostic Compositionality
This repository contains RMarkdown with code to analyse sound similarity in novel vocalizations..
Implementing a transformer model for the SCAN compositionality tasks.