# Improving Zero Shot Learning Baselines with Commonsense Knowledge

arXiv: https://arxiv.org/abs/2012.06236 · HTML: https://mlsanity.com/paper/2012.06236/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Dec 2020.

## Built on by (most cited first, 10 of 10)

- [Knowledge-aware Zero-Shot Learning: Survey and Perspective](https://mlsanity.com/paper/2103.00070/) (2021)
- [Zero-shot and Few-shot Learning with Knowledge Graphs: A Comprehensive Survey](https://mlsanity.com/paper/2112.10006/) (2021)
- [Disentangled Ontology Embedding for Zero-shot Learning](https://mlsanity.com/paper/2206.03739/) (2022)
- [A Survey on Visual Transfer Learning using Knowledge Graphs](https://mlsanity.com/paper/2201.11794/) (2022)
- [Benchmarking Knowledge-driven Zero-shot Learning](https://mlsanity.com/paper/2106.15047/) (2021)
- [Geometry-Aware Adaptation for Pretrained Models](https://mlsanity.com/paper/2307.12226/) (2023)
- [Open-Pose 3D Zero-Shot Learning: Benchmark and Challenges](https://mlsanity.com/paper/2312.07039/) (2023)
- [Enhancing Explainability in Multimodal Large Language Models Using Ontological Context](https://mlsanity.com/paper/2409.18753/) (2024)
- [Zero-Shot Learning in Named-Entity Recognition with External Knowledge](https://mlsanity.com/paper/2111.07734/) (2021)
- [A Probabilistic-Logic based Commonsense Representation Framework for Modelling Inferences with Multiple Antecedents and Varying Likelihoods](https://mlsanity.com/paper/2211.16822/) (2022)

## Abstract (arXiv)

Zero shot learning -- the problem of training and testing on a completely disjoint set of classes -- relies greatly on its ability to transfer knowledge from train classes to test classes. Traditionally semantic embeddings consisting of human defined attributes (HA) or distributed word embeddings (DWE) are used to facilitate this transfer by improving the association between visual and semantic embeddings. In this paper, we take advantage of explicit relations between nodes defined in ConceptNet, a commonsense knowledge graph, to generate commonsense embeddings of the class labels by using a graph convolution network-based autoencoder. Our experiments performed on three standard benchmark datasets surpass the strong baselines when we fuse our commonsense embeddings with existing semantic embeddings i.e. HA and DWE.