# Enhancing Multi-Image Question Answering via Submodular Subset Selection

Source: [Enhancing Multi-Image Question Answering via Submodular Subset Selection](<https://www.sota2.com/research/paper/68fa789d1e209131cea79572>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T04:42:14.951Z

Authors: Aaryan Sharma, Shivansh Gupta, Samar Agarwal, Vishak Prasad C., Ganesh Ramakrishnan

Published: 2025-05-15T17:41:52.000Z

## Abstract

Large multimodal models (LMMs) have achieved high performance in vision-language tasks involving single image but they struggle when presented with a collection of multiple images (Multiple Image Question Answering scenario). These tasks, which involve reasoning over large number of images, present issues in scalability (with increasing number of images) and retrieval performance. In this work, we propose an enhancement for retriever framework introduced in MIRAGE model using submodular subset selection techniques. Our method leverages query-aware submodular functions, such as GraphCut, to pre-select a subset of semantically relevant images before main retrieval component. We demonstrate that using anchor-based queries and augmenting the data improves submodular-retriever pipeline effectiveness, particularly in large haystack sizes.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2505.10533>)

[PDF](<https://arxiv.org/pdf/2505.10533.pdf>)