Woodpecker: Hallucination Correction for Multimodal Large Language Models

TL;DR

Woodpecker is a training-free, post-processing method designed to detect and correct hallucinations in multimodal large language models by analyzing and refining generated text based on visual content.

Contribution

It introduces a novel, training-free approach for hallucination correction in MLLMs that operates through a five-stage process without model retraining.

Findings

30.66% accuracy improvement on POPE benchmark for MiniGPT-4

24.33% accuracy improvement on POPE benchmark for mPLUG-Owl

Effective in both quantitative and qualitative evaluations

Abstract

Hallucination is a big shadow hanging over the rapidly evolving Multimodal Large Language Models (MLLMs), referring to the phenomenon that the generated text is inconsistent with the image content. In order to mitigate hallucinations, existing studies mainly resort to an instruction-tuning manner that requires retraining the models with specific data. In this paper, we pave a different way, introducing a training-free method named Woodpecker. Like a woodpecker heals trees, it picks out and corrects hallucinations from the generated text. Concretely, Woodpecker consists of five stages: key concept extraction, question formulation, visual knowledge validation, visual claim generation, and hallucination correction. Implemented in a post-remedy manner, Woodpecker can easily serve different MLLMs, while being interpretable by accessing intermediate outputs of the five stages. We evaluate Woodpecker both quantitatively and qualitatively and show the huge potential of this new paradigm. On the POPE benchmark, our method obtains a 30.66%/24.33% improvement in accuracy over the baseline MiniGPT-4/mPLUG-Owl. The source code is released at https://github.com/BradyFU/Woodpecker.