FilterRAG: Zero-Shot Informed Retrieval-Augmented Generation to Mitigate Hallucinations in VQA

TL;DR

FilterRAG is a retrieval-augmented framework that significantly reduces hallucinations and enhances robustness in Visual Question Answering by grounding answers in external knowledge sources.

Contribution

It introduces FilterRAG, a novel approach combining BLIP-VQA with retrieval-augmented generation to improve factual accuracy in VQA models.

Findings

Achieves 36.5% accuracy on OK-VQA dataset.

Reduces hallucinations in knowledge-driven VQA.

Improves robustness in Out-of-Distribution scenarios.

Abstract

Visual Question Answering requires models to generate accurate answers by integrating visual and textual understanding. However, VQA models still struggle with hallucinations, producing convincing but incorrect answers, particularly in knowledge-driven and Out-of-Distribution scenarios. We introduce FilterRAG, a retrieval-augmented framework that combines BLIP-VQA with Retrieval-Augmented Generation to ground answers in external knowledge sources like Wikipedia and DBpedia. FilterRAG achieves 36.5% accuracy on the OK-VQA dataset, demonstrating its effectiveness in reducing hallucinations and improving robustness in both in-domain and Out-of-Distribution settings. These findings highlight the potential of FilterRAG to improve Visual Question Answering systems for real-world deployment.