CutPaste&Find: Efficient Multimodal Hallucination Detector with Visual-aid Knowledge Base
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911091169165312 |
|---|---|
| author | Nguyen, Cong-Duy Wu, Xiaobao Vu, Duc Anh Zhao, Shuai Nguyen, Thong Luu, Anh Tuan |
| author_facet | Nguyen, Cong-Duy Wu, Xiaobao Vu, Duc Anh Zhao, Shuai Nguyen, Thong Luu, Anh Tuan |
| contents | Large Vision-Language Models (LVLMs) have demonstrated impressive multimodal reasoning capabilities, but they remain susceptible to hallucination, particularly object hallucination where non-existent objects or incorrect attributes are fabricated in generated descriptions. Existing detection methods achieve strong performance but rely heavily on expensive API calls and iterative LVLM-based validation, making them impractical for large-scale or offline use. To address these limitations, we propose CutPaste\&Find, a lightweight and training-free framework for detecting hallucinations in LVLM-generated outputs. Our approach leverages off-the-shelf visual and linguistic modules to perform multi-step verification efficiently without requiring LVLM inference. At the core of our framework is a Visual-aid Knowledge Base that encodes rich entity-attribute relationships and associated image representations. We introduce a scaling factor to refine similarity scores, mitigating the issue of suboptimal alignment values even for ground-truth image-text pairs. Comprehensive evaluations on benchmark datasets, including POPE and R-Bench, demonstrate that CutPaste\&Find achieves competitive hallucination detection performance while being significantly more efficient and cost-effective than previous methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_12591 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CutPaste&Find: Efficient Multimodal Hallucination Detector with Visual-aid Knowledge Base Nguyen, Cong-Duy Wu, Xiaobao Vu, Duc Anh Zhao, Shuai Nguyen, Thong Luu, Anh Tuan Computer Vision and Pattern Recognition Computation and Language Large Vision-Language Models (LVLMs) have demonstrated impressive multimodal reasoning capabilities, but they remain susceptible to hallucination, particularly object hallucination where non-existent objects or incorrect attributes are fabricated in generated descriptions. Existing detection methods achieve strong performance but rely heavily on expensive API calls and iterative LVLM-based validation, making them impractical for large-scale or offline use. To address these limitations, we propose CutPaste\&Find, a lightweight and training-free framework for detecting hallucinations in LVLM-generated outputs. Our approach leverages off-the-shelf visual and linguistic modules to perform multi-step verification efficiently without requiring LVLM inference. At the core of our framework is a Visual-aid Knowledge Base that encodes rich entity-attribute relationships and associated image representations. We introduce a scaling factor to refine similarity scores, mitigating the issue of suboptimal alignment values even for ground-truth image-text pairs. Comprehensive evaluations on benchmark datasets, including POPE and R-Bench, demonstrate that CutPaste\&Find achieves competitive hallucination detection performance while being significantly more efficient and cost-effective than previous methods. |
| title | CutPaste&Find: Efficient Multimodal Hallucination Detector with Visual-aid Knowledge Base |
| topic | Computer Vision and Pattern Recognition Computation and Language |
| url | https://arxiv.org/abs/2502.12591 |