Shifting the Baseline: Single Modality Performance on Visual Navigation & QA

Jesse Thomason, Daniel Gordon, Yonatan Bisk


Abstract
We demonstrate the surprising strength of unimodal baselines in multimodal domains, and make concrete recommendations for best practices in future research. Where existing work often compares against random or majority class baselines, we argue that unimodal approaches better capture and reflect dataset biases and therefore provide an important comparison when assessing the performance of multimodal techniques. We present unimodal ablations on three recent datasets in visual navigation and QA, seeing an up to 29% absolute gain in performance over published baselines.
Anthology ID:
N19-1197
Volume:
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
Month:
June
Year:
2019
Address:
Minneapolis, Minnesota
Editors:
Jill Burstein, Christy Doran, Thamar Solorio
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1977–1983
Language:
URL:
https://aclanthology.org/N19-1197
DOI:
10.18653/v1/N19-1197
Bibkey:
Cite (ACL):
Jesse Thomason, Daniel Gordon, and Yonatan Bisk. 2019. Shifting the Baseline: Single Modality Performance on Visual Navigation & QA. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 1977–1983, Minneapolis, Minnesota. Association for Computational Linguistics.
Cite (Informal):
Shifting the Baseline: Single Modality Performance on Visual Navigation & QA (Thomason et al., NAACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/N19-1197.pdf
Data
EQAIQUAD