A Small but Informed and Diverse Model: The Case of the Multimodal GuessWhat!? Guessing Game

Claudio Greco, Alberto Testoni, Raffaella Bernardi, Stella Frank


Abstract
Pre-trained Vision and Language Transformers achieve high performance on downstream tasks due to their ability to transfer representational knowledge accumulated during pretraining on substantial amounts of data. In this paper, we ask whether it is possible to compete with such models using features based on transferred (pre-trained, frozen) representations combined with a lightweight architecture. We take a multimodal guessing task as our testbed, GuessWhat?!. An ensemble of our lightweight model matches the performance of the finetuned pre-trained transformer (LXMERT). An uncertainty analysis of our ensemble shows that the lightweight transferred representations close the data uncertainty gap with LXMERT, while retaining model diversity leading to ensemble boost. We further demonstrate that LXMERT’s performance gain is due solely to its extra V&L pretraining rather than because of architectural improvements. These results argue for flexible integration of multiple features and lightweight models as a viable alternative to large, cumbersome, pre-trained models.
Anthology ID:
2022.clasp-1.1
Volume:
Proceedings of the 2022 CLASP Conference on (Dis)embodiment
Month:
September
Year:
2022
Address:
Gothenburg, Sweden
Venue:
CLASP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1–10
Language:
URL:
https://aclanthology.org/2022.clasp-1.1
DOI:
Bibkey:
Cite (ACL):
Claudio Greco, Alberto Testoni, Raffaella Bernardi, and Stella Frank. 2022. A Small but Informed and Diverse Model: The Case of the Multimodal GuessWhat!? Guessing Game. In Proceedings of the 2022 CLASP Conference on (Dis)embodiment, pages 1–10, Gothenburg, Sweden. Association for Computational Linguistics.
Cite (Informal):
A Small but Informed and Diverse Model: The Case of the Multimodal GuessWhat!? Guessing Game (Greco et al., CLASP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.clasp-1.1.pdf
Data
COCOGuessWhat?!Visual Genome