Woodpecker: hallucination correction for multimodal large language models
Science China Information Sciences · 2024-12 · 67 (12)
DOI: 10.1007/s11432-024-4251-x ↗Kurz gefasst
Fachartikel in der Zeitschrift Science China Information Sciences, online erschienen am 13.12.2024. Verfasst von Shukang Yin und 9 weiteren Autor:innen. Die Arbeit wurde bisher 118-mal zitiert (Crossref) und nennt 53 Literaturangaben.
Abstract (Originalsprache)
Hallucinations is a big shadow hanging over the rapidly evolving multimodal large language models (MLLMs), referring to that the generated text is inconsistent with the image content. 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 woodpeckers heal 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.
Zitieren
APA
Yin, S., Fu, C., Zhao, S., Xu, T., Wang, H., Sui, D., Shen, Y., Li, K., Sun, X., & Chen, E. (2024). Woodpecker: hallucination correction for multimodal large language models. Science China Information Sciences, 67(12). https://doi.org/10.1007/s11432-024-4251-x
Vancouver
Yin S, Fu C, Zhao S, Xu T, Wang H, Sui D, et al. Woodpecker: hallucination correction for multimodal large language models. Science China Information Sciences. 2024;67(12). doi:10.1007/s11432-024-4251-x
BibTeX
@article{Yin2024,
author = {Yin, Shukang and Fu, Chaoyou and Zhao, Sirui and Xu, Tong and Wang, Hao and Sui, Dianbo and Shen, Yunhang and Li, Ke and Sun, Xing and Chen, Enhong},
title = {Woodpecker: hallucination correction for multimodal large language models},
journal = {Science China Information Sciences},
year = {2024},
volume = {67},
number = {12},
doi = {10.1007/s11432-024-4251-x}
}
EUVIMED