GRILL: Grounded Vision-language Pre-training via Aligning Text and Image Regions

TL;DR

GRILL is a novel vision-language model that learns object grounding through alignments, enabling it to excel in diverse zero-/few-shot tasks like VQA, captioning, and grounding, surpassing existing methods.

Contribution

The paper introduces GRILL, a model that effectively generalizes to multiple vision-language tasks with minimal training data by leveraging object-text alignment for grounding.

Findings

Outperforms state-of-the-art few-shot methods on various VL tasks.

Enables zero-/few-shot transfer for grounding tasks.

Demonstrates strong generalization across diverse VL tasks.

Abstract

Generalization to unseen tasks is an important ability for few-shot learners to achieve better zero-/few-shot performance on diverse tasks. However, such generalization to vision-language tasks including grounding and generation tasks has been under-explored; existing few-shot VL models struggle to handle tasks that involve object grounding and multiple images such as visual commonsense reasoning or NLVR2. In this paper, we introduce GRILL, GRounded vIsion Language aLigning, a novel VL model that can be generalized to diverse tasks including visual question answering, captioning, and grounding tasks with no or very few training instances. Specifically, GRILL learns object grounding and localization by exploiting object-text alignments, which enables it to transfer to grounding tasks in a zero-/few-shot fashion. We evaluate our model on various zero-/few-shot VL tasks and show that it consistently surpasses the state-of-the-art few-shot methods.