EaSe: A Diagnostic Tool for VQA Based on Answer Diversity
| Autor: | Jolly, S., Pezzelle, S., Nabi, M., Toutanova, K., Rumshisky, A., Zettlemoyer, L., Hakkani-Tur, D., Beltagy, I., Bethard, S., Cotterell, R., Chakraborty, T., Zhou, Y. |
|---|---|
| Přispěvatelé: | Language and Computation (ILLC, FNWI/FGw), ILLC (FNWI) |
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Information retrieval
Computer science 0202 electrical engineering electronic engineering information engineering Question answering 020201 artificial intelligence & image processing Sample (statistics) 02 engineering and technology 010501 environmental sciences Entropy (energy dispersal) 01 natural sciences 0105 earth and related environmental sciences |
| Zdroj: | The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: NAACL-HLT 2021 : proceedings of the conference : June 6-11, 2021, 2407-2414 STARTPAGE=2407;ENDPAGE=2414;TITLE=The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies NAACL-HLT |
| DOI: | 10.18653/v1/2021.naacl-main.192 |
| Popis: | We propose EASE, a simple diagnostic tool for Visual Question Answering (VQA) which quantifies the difficulty of an image, question sample. EASE is based on the pattern of answers provided by multiple annotators to a given question. In particular, it considers two aspects of the answers: (i) their Entropy; (ii) their Semantic content. First, we prove the validity of our diagnostic to identify samples that are easy/hard for state-of-art VQA models. Second, we show that EASE can be successfully used to select the most-informative samples for training/fine-tuning. Crucially, only information that is readily available in any VQA dataset is used to compute its scores. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|