Black-box Model Ensembling for Textual and Visual Question Answering via Information Fusion

TL;DR

This paper introduces InfoSel, a data-efficient ensemble method that dynamically selects the best black-box model for textual and visual question answering, significantly improving accuracy with minimal training data.

Contribution

The paper presents a novel ensemble approach that does not require access to model probabilities, enabling effective combination of black-box models for QA tasks.

Findings

Achieves up to +5.19% F1-score improvement

Operates effectively with only 1K training instances

Applicable to both textual and multimodal QA tasks

Abstract

A diverse range of large language models (LLMs), e.g., ChatGPT, and visual question answering (VQA) models, e.g., BLIP, have been developed for solving textual and visual question answering tasks. However, fine-tuning these models is either difficult, as it requires access via APIs, rendering them as black-boxes, or costly due to the need of tuning a large number of parameters. To address this, we introduce InfoSel, a data-efficient ensemble method that learns to dynamically pick the winner from existing black-box models for predictions on both textual and multimodal visual question answering tasks. Unlike traditional ensemble models, InfoSel does not rely on prediction probabilities or confidences, which typically are not available in black-box models. Experimental results on four datasets demonstrate that our approach achieves an absolute increase of up to +5.19\% in the F1-score compared to standalone LLMs using only 1K training instances.