Hi there! I am Anda Cheng (程安达), currently serving as an Algorithm Expert at Ant Group. My research interests encompass LLMs, AI agents, and trustworthy AI. I received my Ph.D. from the Institute of Automation, Chinese Academy of Sciences (CASIA) in 2023, where I was fortunate to be advised by Professor Jian Cheng. Prior to that, I obtained my B.Eng. degree from Northeastern University in 2017.

Work Experience

• Research Intern, July 2022 - Jan. 2023, Alibaba DAMO Academy.
• Staff, July 2023 - present, Ant Group.

Publications

• [ICLR’26] Wei Huang*, Anda Cheng*(co-first author), Yinggui Wang. GradPruner: Gradient-Guided Layer Pruning Enabling Efficient Fine-Tuning and Inference for LLMs.
• [VLDB’26] Wei Huang*, Anda Cheng*(co-first author, corresponding author), Yinggui Wang, Lei Wang, Tao Wei. LLM-AutoDP: Automatic Data Processing via LLM Agents for Model Fine-tuning.
• [EMNLP’25 Main] Anda Cheng, Wei Huang, Yinggui Wang. A Fully Probabilistic Perspective on Large Language Model Unlearning: Evaluation and Optimization.
• [EMNLP’25 Main] Wei Huang*, Anda Cheng*(co-first author), Yinggui Wang. Mitigating Catastrophic Forgetting in Large Language Models with Forgetting-aware Pruning.
• [EMNLP’25 Finding] Wei Huang*, Anda Cheng*(co-first author), Zhao Zhang, Yinggui Wang. DPF-CM: A Data Processing Framework with Privacy-Preserving Vector Databases for Chinese Medical LLMs Training and Deployment.
• [IJCAI’25] Yiran Xu, Nan Zhong, Guobiao Li, Anda Cheng, Yinggui Wang, Zhenxing Qian, Xinpeng Zhang. Fine-grained Prompt Screening: Defending Against Backdoor Attack on Text-to-Image Diffusion Models.
• [arxiv’25] Chunyang Li, Zifeng Kang, Junwei Zhang, Zhuo Ma, Anda Cheng, Xinghua Li, Jianfeng Ma. The Shawshank Redemption of Embodied AI: Understanding and Benchmarking Indirect Environmental Jailbreaks.
• [arxiv’25] Chunyang Li, Junwei Zhang, Anda Cheng, Zhuo Ma, Xinghua Li, Jianfeng Ma. CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models.
• [ICASSP’24] Wei Huang, Yinggui Wang, Anda Cheng, Aihui Zhou, Chaofan Yu, Lei Wang. A Fast, Performant, Secure Distributed Training Framework For LLM.
• [arxiv’24] Zhipeng Wan, Anda Cheng, Yinggui Wang, Lei Wang: Information Leakage from Embedding in Large Language Models.
• [ICASSP’23, oral] Anda Cheng, Jian Cheng: APGP: Accuracy-Preserving Generative Perturbation for Defending Against Model Cloning Attacks.
• [arxiv’23] Anda Cheng, Zhen Wang, Yaliang Li, Jian Cheng: HPN: Personalized Federated Hyperparameter Optimization.
• [AAAI’22] Anda Cheng, Jiaxing Wang, Xi Sheryl Zhang, Qiang Chen, Peisong Wang, Jian Cheng. DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy.
• [CVPR’22] Anda Cheng, Peisong Wang, Xi Sheryl Zhang, Jian Cheng: Differentially Private Federated Learning with Local Regularization and Sparsification.
• [NeurIPS’22] Weihan Cao, Yifan Zhang, Jianfei Gao, Anda Cheng, Ke Cheng, Jian Cheng. PKD: General Distillation Framework for Object Detectors via Pearson Correlation Coefficient.
• [TCSVT’20] Qiang Chen, Anda Cheng, Xiangyu He, Peisong Wang, Jian Cheng. SpatialFlow: Bridging All Tasks for Panoptic Segmentation.
• [TNNLS’20] Peisong Wang, Xiangyu He, Qiang Chen, Anda Cheng, Qingshan Liu, Jian Cheng. Unsupervised Network Quantization via Fixed-Point Factorization.
• [PR’20] Qiang Chen, Peisong Wang, Anda Cheng, Wanguo Wang, Yifan Zhang, Jian Cheng. Robust one-stage object detection with location-aware classifiers.

Academic Activity

• Reviewer: ICML, NeurIPS, ICLR, ACL, EMNLP, CVPR, ECCV, AAAI, AISTATS, ICASSP.