Zeroth-Order Optimization is Secretly Single-Step Policy Optimization. Junbin Qiu, Zhengpeng Xie, Xiangda Yan, Yongjie Yang, Yao Shu# Key Words: Zeroth-Order Optimization, Policy Optimization, Variance Reduction, Query Reuse Workshop on Tiny Titans: The next wave of On-Device Learning for Foundational Models @ ICML, 2025 [OpenReview]
ReDit: Reward Dithering for Improved LLM Policy Optimization. Chenxing Wei, Jiarui Yu, Ying Tiffany He, Hande Dong, Yao Shu#, Fei Yu Key Words: Reinforcement Learning with Verifiable Reward, Noisy Reward, Large Language Models Workshop on Models of Human Feedback for AI Alignment @ ICML, 2025 [OpenReview]
Optimization and Robustness-Informed Membership Inference Attacks for LLMs. Zichen Song, Qixin Zhang, Ming Li, Yao Shu# Key Words: Membership Inference Attack, Large Language Models Workshop on Data in Generative Models (The Bad, the Ugly, and the Greats) @ ICML, 2025 [OpenReview]
Flexora: Flexible Low Rank Adaptation for Large Language Models. Chenxing Wei*, Yao Shu*, Ying Tiffany He, Fei Richard Yu Key Words: Layer Selection, Parameter-Efficient Fine-Tuning, Large Language Models Workshop on Fine-Tuning in Modern Machine Learning @ NeurIPS, 2024 In The 63rd Annual Meeting of the Association for Computational Linguistics (ACL), 2025 [HuggingFace, arXiv, code]
Refining Adaptive Zeroth-Order Optimization at Ease. Yao Shu, Qixin Zhang, Kun He, Zhongxiang Dai Key Words: Zeroth-Order Optimization, Adaptive Method, Variance Reduction In The 42nd International Conference on Machine Learning (ICML), 2025 [OpenReview, arXiv]
Ferret: Federated Full-Parameter Tuning at Scale for Large Language Models. Yao Shu*#, Wenyang Hu*, See-Kiong Ng, Bryan Kian Hsiang Low, Fei Richard Yu Key Words: Large Language Models, Federated Full-Parameter Tuning Workshop on Federated Foundation Models @ NeurIPS (Oral), 2024 In The 42nd International Conference on Machine Learning (ICML), 2025 [HuggingFace, OpenReview, arXiv, code]
Multinoulli Extension: A Lossless Yet Effective Probabilistic Framework for Subset Selection over Partition Constraints. Qixin Zhang, Wei Huang, Can Jin, Puning Zhao, Yao Shu, Li Shen, Dacheng Tao Key Words: Subset Selection In The 42nd International Conference on Machine Learning (ICML), 2025 [OpenReview]
WMarkGPT: Watermarked Image Understanding via Multimodal Large Language Models. Songbai Tan, Xuerui Qiu, Yao Shu, Gang Xu, Linrui Xu, Xiangyu Xu, Huiping Zhuang, Ming Li, Fei Richard Yu Key Words: Watermarked Image Understanding, Multimodal Large Language Models In The 42nd International Conference on Machine Learning (ICML), 2025 [OpenReview]
Meta-Prompt Optimization for LLM-Based Sequential Decision Making. Mingze Kong, Zhiyong Wang, Yao Shu, Zhongxiang Dai Key Words: Large Language Models, Decision-Making Workshop on Reasoning and Planning for Large Language Models @ ICLR, 2025 [arXiv]
FSL-Rectifier: Rectify Outliers in Few-Shot Learning via Test-Time Augmentation. Yunwei Bai, Ying Kiat Tan, Shiming Chen, Yao Shu, Tsuhan Chen Key Words: Test-Time Augmentation, Few-Shot Learning In The 39th Annual AAAI Conference on Artificial Intelligence (AAAI), 2025 [arXiv, OpenReview]
PAFT: Prompt-Agnostic Fine-Tuning. **Chenxing Wei, Yao Shu, Mingwen Ou, Ying Tiffany He, Fei Richard Yu Key Words: Prompt-Agnostic, Fine-Tuning, Robustness [arXiv, HuggingFace, code]
Localized Zeroth-Order Prompt Optimization. Wenyang Hu*, Yao Shu*****, Zongmin Yu, Zhaoxuan Wu, Xiangqiang Lin, Zhongxiang Dai, See-Kiong Ng, Bryan Kian Hsiang Low Key Words: Prompt Optimization, Zeroth-Order Optimization, Neural Tangent Kernel, Large Language Models Workshop on In-Context Learning @ ICML, 2024 In The 38th Conference on Neural Information Processing Systems (NeurIPS Spotlight), 2024 [arXiv, OpenReview, code]
Prompt Optimization with EASE? Efficient Ordering-aware Automated Selection of Exemplars. Zhaoxuan Wu, Xiaoqiang Lin, Zhongxiang Dai, Wenyang Hu, Yao Shu, See-Kiong Ng, Patrick Jaillet, Bryan Kian Hsiang Low Key Words: Prompt Optimization, Neural Bandit, Large Language Models Workshop on In-Context Learning @ ICML, 2024 In The 38th Conference on Neural Information Processing Systems (NeurIPS), 2024 [arXiv, OpenReview, code]
OptEx: Expediting First-Order Optimization with Approximately Parallelized Iterations. Yao Shu#, Jiongfeng Fang, Ying Tiffany He, Fei Richard Yu Key Words: First-Order Optimization, Parallelization In The 38th Conference on Neural Information Processing Systems (NeurIPS), 2024 [arXiv, OpenReview, code]