Banghao Chi

Banghao Chi

Intern @ NCSA | CA @ UIUC

Publications

You can also find my articles on my Google Scholar profile.

Journal Articles

Q-LiDAR: Efficient and Accurate Training-Free Quantization for Point Cloud 3D Object Detection Models

Published in Conference on Computer Vision and Pattern Recognition (CVPR), 2024

This paper presents novel dynamic and static post-training quantization techniques for 3D object detection models, achieving 35% reduction in inference time while maintaining accuracy through progressive quantization and sensitivity analysis.

Recommended citation: Chi, B., Zheng, H., & Zhang, M. (2024). "Q-LiDAR: Efficient and Accurate Training-Free Quantization for Point Cloud 3D Object Detection Models." Conference on Computer Vision and Pattern Recognition (CVPR). Submitted.
Download Paper

LLMarking: An automatic short-answer grading system based on large language model

Published in Educational Advances in Artificial Intelligence (EAAI), 2024

This paper introduces an automatic short-answer grading system utilizing large language models, achieving F1 scores of 90.5% and 86.1% on computer science and finance datasets respectively through innovative prompt engineering and model fine-tuning.

Recommended citation: Wang, H., Chi, B., Wu, Y., et al. (2024). "LLMarking: An automatic short-answer grading system based on large language model." 15th Symposium on Educational Advances in Artificial Intelligence (EAAI). Submitted.
Download Paper

Research Advanced in the Object Detection Based on Deep Learning

Published in 2022 International Conference on Applied Physics and Computing (ICAPC), 2022

This paper provides a comprehensive evaluation of object detection research advances, categorizing algorithms into traditional detection algorithms, anchor-based and anchor-free detection algorithms, with experimental analysis on common datasets.

Recommended citation: Chi, B. (2022). "Research Advanced in the Object Detection Based on Deep Learning." 2022 International Conference on Applied Physics and Computing (ICAPC). DOI: 10.1109/ICAPC57304.2022.00092
Download Paper