General & Reliable Intelligence Lab (GRINLab), led by Dr. Chi Liu, is dedicated to advancing reliable, responsible, and ethical AI. Our research spans AI-generated content safety, AI attack and defense, data privacy, as well as multimodal decision-making for healthcare, medical AI ethics, clinical decision support, and visual cognition. With AI for Social Good as our ultimate mission, we deliver rigorous research that empowers intelligent technologies to be safer, more inclusive, and more human-centered, turning academic insights into real-world impacts that help society become good, better, and best
Research Focus Keywords
GRINLab 聚焦人工智能的可靠性、安全性与医疗应用，围绕可信内容生成、智能系统防护、多模态临床决策以及医学 AI 伦理等关键问题开展研究。
Research Focus
Reliable · Secure · Healthcare
聚焦 AI 生成内容的可信识别、风险治理与版权保护，研究深度伪造内容的检测和溯源方法，降低大模型幻觉，并推动模型与人类认知、价值观和偏好对齐。同时关注去歧视、NSFW 内容过滤、数字水印及生成内容版权确权等关键问题。
面向人工智能系统在真实环境中的安全威胁，研究后门攻击、对抗攻击、数据投毒及越狱攻击的机理、检测与防御方法。通过构建可验证、可监测并具备主动防护能力的模型与系统，提升 AI 在训练、部署和交互全过程中的安全性与鲁棒性。
融合医学影像、文本、检验指标等多模态信息，构建疾病诊断、临床决策支持与报告问答系统，并探索医疗基础模型的微调、部署与应用，提升诊疗过程的智能化、准确性与可解释性。同时关注视觉认知行为、眼脑一体化等前沿方向。
面向医学人工智能落地过程中的伦理、安全与可靠性挑战，研究医疗数据隐私保护和数据质量、跨机构与跨域泛化、环境鲁棒性、算法去偏和公平性保障、可及性及经济性等问题，推动医疗 AI 在不同人群、设备和临床环境中的稳定、公平与可信应用。
论文录用、荣誉奖项、学术报告与团队生活。
GRINLab welcomes motivated students with a genuine passion for research. Explore our postgraduate, undergraduate research trainee, and external volunteer opportunities.
Congratulations to Mengna Sun (Caroline), Yaorong Zhong, Yinghe Zhang, and Pengyu Gao on presenting their papers at BlockSys 2026. Special congratulations to Pengyu Gao for receiving the Best Poster Award.
Congratulations to undergraduate research trainee Yuzhuo Zhou on the publication of “Fundus Image-Based Glaucoma Screening via Retinal Knowledge-Oriented Dynamic Multi-Level Feature Integration” in the JCR Q1 journal Knowledge-Based Systems.
Congratulations to Pengyu Gao, Zhuo Luo, Yaorong Zhong, Hui Xu, and Mengna Sun on receiving their master’s degrees. We wish them continued success in their future academic and professional endeavors.
Congratulations to Yinghe Zhang, Jincheng Liu, Yunfei Qi, and Yuting Huang on receiving their bachelor’s degrees. We wish them continued success in the next chapter of their journeys.
Congratulations to Yuting Huang on being selected as the Faculty’s Outstanding Graduate Representative and receiving the Outstanding Graduation Project Award.
GRINLab contributed to the collaborative study “Cost-Effectiveness Analysis of Human–AI Collaboration in Disease Screening,” published in Nature Health. Congratulations to all collaborators on this achievement at the intersection of medical AI and health economics.
Congratulations to graduate researcher Hui Xu on the acceptance of “Causal Fingerprints of AI Generative Models” at ICASSP 2026, a flagship conference in signal processing (CCF-B).
Congratulations to undergraduate research trainee Liwen Yu on the acceptance of “Learning to Look before Learning to Like: Incorporating Human Visual Cognition into Aesthetic Quality Assessment” for poster presentation at CogSci 2026 (CCF-B).
Congratulations to undergraduate research trainee Jincheng Liu on the publication of “Rethinking Bias in Generative Data Augmentation for Medical AI: A Frequency Recalibration Method” in the proceedings of AAAI 2026 (CCF-A).
我们是一支充满活力、相互支持的研究团队。
PhD student @ CityU Macau.
研究兴趣
图像增强，AI伦理
PhD student @ CityU Macau.
方班研究兴趣
AIGC Image
PhD student @ Monash University.
Co-supervised
PhD student @ CityU Macau.
方班研究兴趣
AIGC内容安全