EaSe: A Diagnostic Tool for VQA Based on Answer Diversity
| Authors |
|
|---|---|
| Publication date | 2021 |
| Host editors |
|
| Book title | The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies |
| Book subtitle | NAACL-HLT 2021 : proceedings of the conference : June 6-11, 2021 |
| ISBN (electronic) |
|
| Event | 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021 |
| Pages (from-to) | 2407-2414 |
| Number of pages | 8 |
| Publisher | Stroudsburg, PA: The Association for Computational Linguistics |
| Organisations |
|
| Abstract |
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.
|
| Document type | Conference contribution |
| Note | With supplementary data |
| Language | English |
| Published at | https://doi.org/10.18653/v1/2021.naacl-main.192 |
| Downloads |
2021.naacl-main.192
(Final published version)
|
| Supplementary materials | |
| Permalink to this page | |
