Probing Image-Language Transformers for Verb Understanding
Citations Over TimeTop 10% of 2021 papers
Abstract
Multimodal image-language transformers have achieved impressive results on a variety of tasks that rely on fine-tuning (e.g., visual question answering and image retrieval). We are interested in shedding light on the quality of their pretrained representationsin particular, if these models can distinguish different types of verbs or if they rely solely on nouns in a given sentence. To do so, we collect a dataset of image-sentence pairs (in English) consisting of 421 verbs that are either visual or commonly found in the pretraining data (i.e., the Conceptual Captions dataset). We use this dataset to evaluate pretrained image-language transformers and find that they fail more in situations that require verb understanding compared to other parts of speech. We also investigate what category of verbs are particularly challenging.
Related Papers
- → Establishing generalized verb-noun instruction-following skills in retarded children(1976)64 cited
- → Deverbal compound comprehension in preschool children(2016)1 cited
- On Verb Replication of Ba-sentence(2009)
- Meaning and Meaning-Extension of the Verb Kata in Kiswahili(2016)
- → Enhancing ESL Writing using Sentence Variety Checklist(2019)