# FVQA 2.0: Introducing Adversarial Samples into Fact-based Visual Question Answering

Source: [FVQA 2.0: Introducing Adversarial Samples into Fact-based Visual Question Answering](<https://www.sota2.com/research/paper/68fa6ed11e209131cea425ef>)

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

Data updated: 2025-10-26T01:07:49.028Z

Authors: Weizhe Lin, Zhilin Wang, Bill Byrne

Published: 2023-03-19T16:07:42.000Z

## Abstract

The widely used Fact-based Visual Question Answering (FVQA) dataset contains visually-grounded questions that require information retrieval using common sense knowledge graphs to answer. It has been observed that the original dataset is highly imbalanced and concentrated on a small portion of its associated knowledge graph. We introduce FVQA 2.0 which contains adversarial variants of test questions to address this imbalance. We show that systems trained with the original FVQA train sets can be vulnerable to adversarial samples and we demonstrate an augmentation scheme to reduce this vulnerability without human annotations.

## 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/2303.10699>)

[PDF](<https://arxiv.org/pdf/2303.10699.pdf>)