ODE: Open-Set Evaluation of Hallucinations in Multimodal Large Language Models
Tu, Yahan, Hu, Rui, Sang, Jitao
–arXiv.org Artificial Intelligence
Hallucination poses a significant challenge for multimodal large language models (MLLMs). However, existing benchmarks for evaluating hallucinations are static, which can lead to potential data contamination. This paper introduces ODE, an open-set, dynamic protocol for evaluating object existence hallucinations in MLLMs. Our framework employs graph structures to model associations between real-word concepts and generates novel samples for both general and domain-specific scenarios. The dynamic combination of concepts, along with various combination principles, ensures a broad sample distribution. Experimental results show that MLLMs exhibit higher hallucination rates with ODE-generated samples, effectively avoiding data contamination. Moreover, these samples can also be used for fine-tuning to improve MLLM performance on existing benchmarks.
arXiv.org Artificial Intelligence
Sep-14-2024
- Country:
- North America
- United States (0.04)
- Mexico > Mexico City
- Mexico City (0.04)
- Asia > China
- North America
- Genre:
- Research Report > New Finding (0.34)
- Technology: