Evidence-Calibrated Multimodal Reasoning for Knowledge-Intensive Visual Question Answering under Perceptual Uncertainty
Main Article Content
Abstract
Visual Question Answering has evolved from simple perceptual recognition tasks to complex knowledge-intensive reasoning challenges. Modern systems are increasingly required to answer queries that cannot be resolved solely from the pixel data present in an image, necessitating the integration of external knowledge bases. However, real-world visual inputs are frequently subject to perceptual uncertainty, encompassing occlusion, poor illumination, adversarial noise, and resolution degradation. Standard multimodal architectures often struggle under such conditions, leading to overconfident predictions grounded in flawed visual interpretations. This paper introduces an Evidence-Calibrated Multimodal Reasoning framework explicitly designed to address Knowledge-Intensive Visual Question Answering under significant perceptual uncertainty. By modeling the output distributions as Dirichlet parameters rather than traditional softmax probabilities, the proposed architecture inherently quantifies both data uncertainty and model uncertainty. We integrate an evidence collection mechanism that dynamically weights the contribution of external knowledge graphs based on the calibrated reliability of the visual and textual modalities. Through extensive evaluation on benchmark datasets, we demonstrate that the proposed framework not only achieves superior accuracy but also provides robust uncertainty estimates that prevent catastrophic failure under severe perceptual degradation. The findings offer a pathway toward more reliable, transparent, and resilient artificial intelligence systems for complex reasoning environments.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in the Journal of the Artificial Intelligence Engineers Consortium (JAIEC) are licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
This license permits unrestricted use, distribution, reproduction, and adaptation in any medium, provided the original author(s) and source are properly credited.
Authors retain copyright of their work while granting the journal the right of first publication.
References
1. Wang, J., Malawade, A. V., Zhou, J., Yu, S. Y., & Al Faruque, M. A. (2024, January). Rs2g: Data-driven scene-graph extraction and embedding for robust autonomous perception and scenario understanding. In 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) (pp. 7478-7487). IEEE.
2. Xia, Y., & Ma, J. (2026). Locality Optimization Refinement with Deformation for Shape Matching via Functional Maps.International Journal of Computer Vision,134(2), 76.
3. Li, B., Zhang, R., Liang, H., Zhang, J., Zhang, J., Chen, X., ... & Wang, J. (2026). Interagent: Physics-based multi-agent command execution via diffusion on interaction graphs. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 15253-15265).
4. Xia, Y., Lu, Y., Gao, Y., & Ma, J. (2024, March). Locality preserving refinement for shape matching with functional maps. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 38, No. 6, pp. 6207-6215).
5. Zhang, W., & Pei, S. (2026). Predictive prefetching for retrieval-augmented generation. Proceedings of Machine Learning Research. The Forty-third International Conference on Machine Learning, ICML 2026. https://par.nsf.gov/biblio/10686293
6. Wang, H., Feng, W., Yu, J., Liu, C., Nie, P., Lin, F., ... & Wei, C. (2026). Search Beyond What Can Be Taught: Evolving the Knowledge Boundary in Agentic Visual Generation.arXiv preprint arXiv:2607.05382.
7. Jing, G., Gao, P., Lee, Y., Hu, Y., & Zhang, H. (2025). 3D-aided pedestrian representation learning for video-based person re-identification. IEEE Transactions on Circuits and Systems for Video Technology, 35 (12), 12830-12845.
8. Liu, C., Ma, L., Zhang, X. T., Zhang, Y., Zhang, H., Yang, X., & Tian, F. (2026). Boosting omni-modal language models: Staged post-training with visually debiased evaluation. arXiv preprint arXiv:2605.12034.
9. Gao, P., Lee, Y., Zhang, X., Chen, Z., & Zhang, H. (2026, March). Remember Me: Bridging the Long-Range Gap in LVLMs with Three-Step Inference-Only Decay Resilience Strategies. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 40, No. 6, pp. 4140-4148).
10. Ding, Y., Li, Z., Huang, D., Li, Z., & Zhang, K. (2022, October). Enhancing multi-view stereo with contrastive matching and weighted focal loss. In 2022 IEEE International Conference on Image Processing (ICIP) (pp. 821-825). IEEE.
11. Mi, L., Wang, W., Tu, W., He, Q., Kong, R., Fang, X., ... & Liu, Y. (2025, March). Empower vision applications with lora lmm. In Proceedings of the Twentieth European Conference on Computer Systems (pp. 261-277).
12. Zhao, H., Wang, Q., Zhan, G., Min, W., Zou, Y., & Cui, S. (2022). Need only one more point (NOOMP): Perspective adaptation crowd counting in complex scenes. IEEE Transactions on Multimedia, 25, 1414-1426.
13. Yang, Y., & Zhang, S. (2024). Pixelwise calibration method for a telecentric structured light system.Applied Optics,63(10), 2562-2569.
14. Qu, W., Wang, J., Gong, Y., Huang, X., & Xiao, L. (2025, June). An end-to-end robust point cloud semantic segmentation network with single-step conditional diffusion models. In 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 27325-27335). IEEE.
15. Yang, Y., & Zhang, S. (2025). Pixel-wise calibration for a multi-focus microscopic 3D imaging system.Optics and Lasers in Engineering,194, 109127.
16. Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of CVPR.
17. Li, Q., Xu, T., Luo, T., Zhong, Y., Li, Y., Zhou, Y., & Hou, C. (2026). Theory-driven label-specific representation for incomplete multi-view multi-label learning. Advances in Neural Information Processing Systems, 38, 121996-122037.
18. Qi, Z., Yuan, Y., Ruan, X., Wang, S., Zhang, W., & Huang, Q. (2024, March). Bias-conflict sample synthesis and adversarial removal debias strategy for temporal sentence grounding in video. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 38, No. 5, pp. 4533-4541).
19. Qu, D., & Ma, Y. (2026). Fourier–Transformer Mixer Network for Efficient Video Scene Graph Prediction.Engineering Proceedings,120(1), 16.
20. Xia, Y., & Ma, J. (2025). Grid-guided sparse Laplacian consensus for robust feature matching.IEEE Transactions on Image Processing,34, 1367-1381.
21. Peng, Qucheng, et al. "3d vision-language gaussian splatting." International Conference on Learning Representations. Vol. 2025. 2025.
22. Li, Q., Luo, T., & Liao, J. (2025). Theory-inspired deep multi-view multi-label learning with incomplete views and noisy labels. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 20706-20715).
23. Qi, Z., Yuan, Y., Ruan, X., Wang, S., Zhang, W., & Huang, Q. (2024). Collaborative debias strategy for temporal sentence grounding in video. IEEE Transactions on Circuits and Systems for Video Technology, 34(11), 10972-10986.
24. Kamal, T., Wenhong, T., Guo, J., Shafiq, M., Shuaihong, J., & Xue, R. (2026). Adaptive Probabilistic Transformer for Medical Image Segmentation. Expert Systems with Applications, 131948.
25. Xia, Y., & Ma, J. (2022). Locality-guided global-preserving optimization for robust feature matching.IEEE Transactions on Image Processing,31, 5093-5108.
26. Huang, H., Zhang, J., Zhang, J., Xu, J., & Wu, Q. (2020). Low-rank pairwise alignment bilinear network for few-shot fine-grained image classification.IEEE Transactions on Multimedia,23, 1666-1680.
27. Jing, G., Gao, P., Hu, Y., Lee, Y., & Zhang, H. (2025, June). ESTI: An Efficient Spatial-Temporal Interaction Network For Video-Based Person Re-Identification. In2025 IEEE International Conference on Multimedia and Expo (ICME)(pp. 1-6). IEEE.
28. Wang, H., Xu, Q., Wang, C., Xue, T., Peng, C., Chen, W., & Lin, F. (2026). Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning.arXiv preprint arXiv:2605.14054.
29. Zhang, Y., Zhang, Y., Zhao, M., Feng, S., & Cheung, Y. M. (2026). How to Ask the AI: A User Perspective Survey for Large Language Model Prompting.IEEE Transactions on Artificial Intelligence.
30. Chen, L., Yang, Y., & Zhang, S. (2024, September). Rapid autofocusing method for digital fringe projection techniques. In Interferometry and Structured Light 2024(Vol. 13135, pp. 92-97). SPIE.
31. Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., et al. (2023). Segment anything. In Proceedings of ICCV.
32. Li, X., Yang, F., Chen, L., & Cai, H. (2016, July). Saliency transfer: An example-based method for salient object detection. In IJCAI (pp. 3411-3417).
33. Wang, Y., Fang, Y., Wang, T., Li, L., Liu, K., Xu, J., ... & Xu, R. (2026). What Limits Vision-and-Language Navigation?.arXiv preprint arXiv:2605.13328.