Quantity Matters: Towards Assessing and Mitigating Number Hallucination in Large Vision-Language Models
This addresses a specific hallucination issue in vision-language models, which is incremental but important for improving model reliability in tasks requiring accurate object counting.
The paper tackles the problem of number hallucination in large vision-language models, where models incorrectly count objects in images, and proposes a training approach that improves performance by 8% over direct finetuning methods.
Large-scale vision-language models have demonstrated impressive skill in handling tasks that involve both areas. Nevertheless, these models frequently experience significant issues with generating inaccurate information, which is hallucination. In this study, we concentrate on a specific type of hallucination-number hallucination, referring to models incorrectly identifying the number of certain objects in pictures. We perform quantitative evaluations regarding number hallucination, showing it to be critical in major open-source large vision-language models. Furthermore, we utilizes two related tasks to conduct an in-depth analysis of number hallucination, revealing the severe inner and outer inconsistency among all tasks. Based on this examination, we devise a training approach aimed at improving consistency to reduce number hallucinations, which leads to an 8% enhancement in performance over direct finetuning methods. Our code and dataset will be released to the community.