Welcome to the SemEval 2023 Shared Task on Visual Word Sense Disambiguation (Visual-WSD)!
Task description: Given a target word, some limited textual context and a set of candidate images, the task is to select the image which corresponds to the intended meaning of the target word. In addition to English, the Visual-WSD test set will contain target and context words in other languages as well (to be announced!).
Trial and training data available! Test languages other than English will be announced on January 2nd, test data will be released on January 10th.
Task website: https://raganato.github.io/vwsd/
Codalab competition: https://codalab.lisn.upsaclay.fr/competitions/8190
Important dates:
* Sample data ready: 1 August 2022
* Training data ready: 18 October 2022
* Additional languages in the test set announced: 2 January 2023
* Evaluation period starts (test set released): 10 January 2023
* Evaluation period ends: 31 January 2023
* Workshop paper submission deadline: February 2023 (tentative)
* SemEval workshop: Summer 2023 (co-located with a major NLP conference)
We encourage many different types of submissions in our shared task. In terms of data, participants are allowed to use (if they wish) the training/trial data set we provide, pretrained vision and language models, as well as other sources of training data (participants will be asked to explain in detail their data sources). In terms of methodology, we would like to encourage novel and exciting research ideas.
We encourage contributions from the following areas (not limited to):
* computer vision,
* natural language processing,
* vision and language,
* (multi-modal) representation learning,
* cognitive sciences,
* machine learning,
* neuro-symbolic learning/reasoning.
Best submissions will be ranked not only in terms of leaderboard performance, but also based on the idea, methodology and analysis. In other words, creative and original research is encouraged!
Task organizers:
* Alessandro Raganato (Department of Informatics, Systems, and Communication, University of Milano-Bicocca, Italy)
* Iacer Calixto (Amsterdam UMC, University of Amsterdam, Netherlands)
* Jose Camacho-Collados (School of Computer Science and Informatics, Cardiff University, United Kingdom)
* Asahi Ushio (School of Computer Science and Informatics, Cardiff University, United Kingdom)
* Mohammad Taher Pilehvar (Tehran Institute for Advanced Studies, Iran)
--
Jose Camacho Collados http://www.josecamachocollados.comhttp://www.josecamachocollados.com/