Abstract
When a model is trying to gather information in an interactive setting, it benefits from asking informative questions. However, in the case of a grounded multi-turn image identification task, previous studies have been constrained to polar yes/no questions (White et al., 2021), limiting how much information the model can gain in a single turn. We present an approach that formulates more informative, open-ended questions. In doing so, we discover that off-the-shelf visual question answering (VQA) models often make presupposition errors, which standard information gain question selection methods fail to account for. To address this issue, we propose a method that can incorporate presupposition handling into both question selection and belief updates. Specifically, we use a two-stage process, where the model first filters out images which are irrelevant to a given question, then updates its beliefs about which image the user intends. Through self-play and human evaluations, we show that our method is successful in asking informative open-ended questions, increasing accuracy over the past state-of-the-art by 14%, while resulting in 48% more efficient games in human evaluations.- Anthology ID:
- 2024.findings-naacl.276
- Volume:
- Findings of the Association for Computational Linguistics: NAACL 2024
- Month:
- June
- Year:
- 2024
- Address:
- Mexico City, Mexico
- Editors:
- Kevin Duh, Helena Gomez, Steven Bethard
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 4429–4442
- Language:
- URL:
- https://aclanthology.org/2024.findings-naacl.276
- DOI:
- Cite (ACL):
- Sedrick Keh, Justin Chiu, and Daniel Fried. 2024. Asking More Informative Questions for Grounded Retrieval. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 4429–4442, Mexico City, Mexico. Association for Computational Linguistics.
- Cite (Informal):
- Asking More Informative Questions for Grounded Retrieval (Keh et al., Findings 2024)
- PDF:
- https://preview.aclanthology.org/naacl24-info/2024.findings-naacl.276.pdf