ReaSCAN is a synthetic navigation task that requires models to reason about surroundings over syntactically difficult languages. (NeurIPS '21)
-
Updated
Nov 28, 2021 - Python
ReaSCAN is a synthetic navigation task that requires models to reason about surroundings over syntactically difficult languages. (NeurIPS '21)
Causal Abstraction of Neural Models Trained to Solve ReaSCAN
The codebase for Inducing Causal Structure for Interpretable Neural Networks
Baby Abstract Reasoning Corpus (BabyARC) dataset engine, for generating grid-world-based abstract reasoning tasks on a large scale.
Distributional Generalization in NLP. A roadmap.
Code from the article: "Lost in Latent Space: Examining Failures of Disentangled Models at Combinatorial Generalisaton" (NeurIPS, 2022)
This repository shares the most important sources used for the reasearch paper "More Diverse Training, Better Compositionality! Evidence from Multimodal Language Learning" by Caspar Volquardsen, Jae Hee Lee, Cornelius Weber, and Stefan Wermter.
Official implementation of ICLR 2023 paper "A Minimalist Dataset for Systematic Generalization of Perception, Syntax, and Semantics"
Diverse Demonstrations Improve In-context Compositional Generalization
ReCOGS: How Incidental Details of a Logical Form Overshadow an Evaluation of Semantic Interpretation
Multiple paper open-source codes of the Microsoft Research Asia DKI group
Object-Centric Disentangled Mechanisms
A curated list of Composable AI methods: Building AI system by composing modules.
Implementation of our paper "Learning to Substitute Span towards Improving Compositional Generalization" @ ACL'2023, Toronto, Canada.
Minimal model of tool discovery and tool innovation using active inference
This repo contains the implementation of the CompMCTG benchmark @ ACL'2024.
Compositional Capabilities of Autoregressive Transformers: A Study on Synthetic, Interpretable Tasks
Official pytorch implementation of CVPR2023 paper "Learning Conditional Attributes for Compositional Zero-Shot Learning"
Living Survey for papers on Compositional Generalization in NLP
Add a description, image, and links to the compositional-generalization topic page so that developers can more easily learn about it.
To associate your repository with the compositional-generalization topic, visit your repo's landing page and select "manage topics."