HMGIE: Hierarchical and Multi-Grained Inconsistency Evaluation for Vision-Language Data Cleansing
Zhu, Zihao, Zhang, Hongbao, Wu, Guanzong, Lyu, Siwei, Wu, Baoyuan
–arXiv.org Artificial Intelligence
Visual-textual inconsistency (VTI) evaluation plays a crucial role in cleansing vision-language data. Its main challenges stem from the high variety of image captioning datasets, where differences in content can create a range of inconsistencies (\eg, inconsistencies in scene, entities, entity attributes, entity numbers, entity interactions). Moreover, variations in caption length can introduce inconsistencies at different levels of granularity as well. To tackle these challenges, we design an adaptive evaluation framework, called Hierarchical and Multi-Grained Inconsistency Evaluation (HMGIE), which can provide multi-grained evaluations covering both accuracy and completeness for various image-caption pairs. Specifically, the HMGIE framework is implemented by three consecutive modules. Firstly, the semantic graph generation module converts the image caption to a semantic graph for building a structural representation of all involved semantic items. Then, the hierarchical inconsistency evaluation module provides a progressive evaluation procedure with a dynamic question-answer generation and evaluation strategy guided by the semantic graph, producing a hierarchical inconsistency evaluation graph (HIEG). Finally, the quantitative evaluation module calculates the accuracy and completeness scores based on the HIEG, followed by a natural language explanation about the detection results. Moreover, to verify the efficacy and flexibility of the proposed framework on handling different image captioning datasets, we construct MVTID, an image-caption dataset with diverse types and granularities of inconsistencies. Extensive experiments on MVTID and other benchmark datasets demonstrate the superior performance of the proposed HMGIE to current state-of-the-art methods.
arXiv.org Artificial Intelligence
Dec-7-2024
- Genre:
- Research Report > Promising Solution (0.34)
- Technology:
- Information Technology
- Artificial Intelligence
- Machine Learning
- Neural Networks > Deep Learning (0.95)
- Performance Analysis > Accuracy (0.69)
- Natural Language
- Chatbot (0.95)
- Large Language Model (1.00)
- Text Processing (1.00)
- Vision (1.00)
- Machine Learning
- Data Science (1.00)
- Artificial Intelligence
- Information Technology