I am currently pursuing a Master of Science in Computer Engineering at the National University of Singapore (NUS). I obtained my undergraduate degree in Artificial Intelligence from the South China University of Technology (SCUT). Currently, I am conducting CEG5003(Research Project) under the guidance of Prof. Xinchao Wang on LLM safety, while also be an intern in the LV Lab supervised by Prof. Shuicheng Yan.
Research Interests: PEFT, AIGC, MLLM, (Visual)Multi-agent System. If you have interest cooperate with me on related topics, please feel free to reach out!
I am looking for summer research internships opportunities in mainland China and the 2027 Spring PhD opportunity, feel free to contact me!
🔥 News
- 2025.11: I submitted 1 paper to CVPR 2026.
- 2025.09: I submitted 2 papers to ICLR 2026.
- 2025.08: 🎉 I join the LV Lab as a research intern, supervised by Prof. Shuicheng Yan.
- 2024.10: I got the No.1 in the Happy Frisbee activity in NUSRI-CQ!
- 2024.09: 🌟 I was selected for my favorite FYP project–Learnable Activation Networks for Vision.
- 2024.07: 🎉🎉 I get the 3+1+1 programme offer from the NUSRI-CQ.
- 2024.03: 🎉 I get the trainee offer from the HSBC Technology China.
- 2024.01: 🎉 I get the excellent group as leader in the winter exchange program in National University of Singapore.
- 2023.12: 🎉 I get the intern offer from the TGAILab, supervised by Prof. Yaochu jin.
📝 Publications
Refinement Provenance Inference: Detecting LLM-Refined Training Prompts from Model Behavior
Bo Yin, Qi Li, Runpeng Yu, Xinchao Wang
We propose an instance-level auditing method that infers, from a model’s behavior, whether it was trained on an original prompt or an LLM-refined version of that prompt within a mixed training corpus.
FeRA: Frequency-Energy Constrained Routing for Effective Diffusion Adaptation Fine-Tuning
Bo Yin, Xiaobin Hu, Xingyu Zhou, Peng-Tao Jiang, Yue Liao, Junwei Zhu, Jiangning Zhang, Ying Tai, Chengjie Wang, Shuicheng Yan
We present a novel fine-tuning approach for diffusion models, termed Frequency-Energy constrained Routing Adaptation (FeRA), which using frequency domain analysis to guide the fine-tuning process, enhancing both efficiency and effectiveness.
Don’t Forget the Nonlinearity: Unlocking Activation Functions in Efficient Fine-Tuning
Bo Yin, Xingyi Yang, Xinchao Wang
We propose a novel and effective fine-tuning paradigm that introduces learnable activation functions to adapt large pre-trained models to downstream tasks while preserving their computational efficiency.
LLacaDM: A Latent Causal Diffusion Model for Multiobjective Reinforcement Learning
Xueming Yan, Bo Yin, Yaochu Jin
We propose a novel multi-objective reinforcement learning framework that leverages latent temporal causal structures to enhance generalization and significantly outperform state-of-the-art baselines in complex dynamic environments.
-
Precise Apple Detection and Localization in Orchards using YOLOv5 for Robotic Harvesting Systems, Jiang Ziyue, Yin Bo, Lu Boyun, AIPMV 2024
-
A Discussion of Migration of Common Neural Network Regularization Methods on SNNs, Lv Yilin, Yin Bo, ISAEECE 2024
📌 Honors
- 2024.09 Xiao Noodles Third-class Scholarship
- 2024.09 SCUT Academic Scholarship
- 2024.05 Tai Hu Innovation Scholarship
- 2022.11 Tai Hu Innovation Scholarship
- 2022.09 SCUT Academic Scholarship
- 2022.09 Huameng Scholarship
📖 Educations
-
National University of Singapore, M.Sc. in Computer Engineering, 2025.09 - 2027.01 -
NUS Research Institute in Chongqing, 3+1+1 Program, 2024.09 - 2025.06
-
South China University of Technology, B.Eng. in Artificial Intelligence, 2021.09 - 2025.07
💻 Internships
-
Huawei, Singapore, 2026.01 - now -
LV Lab, National University of Singapore, Singapore, 2025.08 - now
-
HSBC Technology China, Guangzhou China, 2024.04 - 2024.08 -
TGAILab, Westlake University, Hangzhou China, 2023.12 - 2024.05