BagFormer: Better Cross-Modal Retrieval via bag-wise interaction

TL;DR

BagFormer is a dual encoder model for cross-modal retrieval that uses bag-wise interactions to enhance recall, achieve state-of-the-art performance, and significantly improve efficiency in training and inference.

Contribution

Introduces BagFormer, a dual encoder with bag-wise interactions that balances high retrieval performance with low latency and high throughput.

Findings

Achieves comparable results to state-of-the-art single encoder models.

Reduces latency by 20.72 times.

Increases throughput by 25.74 times.

Abstract

In the field of cross-modal retrieval, single encoder models tend to perform better than dual encoder models, but they suffer from high latency and low throughput. In this paper, we present a dual encoder model called BagFormer that utilizes a cross modal interaction mechanism to improve recall performance without sacrificing latency and throughput. BagFormer achieves this through the use of bag-wise interactions, which allow for the transformation of text to a more appropriate granularity and the incorporation of entity knowledge into the model. Our experiments demonstrate that BagFormer is able to achieve results comparable to state-of-the-art single encoder models in cross-modal retrieval tasks, while also offering efficient training and inference with 20.72 times lower latency and 25.74 times higher throughput.