刘沛羽

学系：大数据科学与技术系

欢迎对高效大语言模型、模型压缩、长上下文推理、智能体数据分析和代码生成等方向感兴趣的同学联系我。如果你有自己的想法，并希望一起做有意义、有影响力的研究，欢迎通过邮件交流。

联系方式：liupeiyustu@163.com

研究兴趣：

研究兴趣围绕自然语言处理与模型压缩展开，近期重点关注大语言模型高效推理、长上下文推理、智能体数据分析、代码生成等方向。

教育背景：

2024年在中国人民大学高瓴人工智能学院获人工智能博士学位，导师为赵鑫教授

工作经历：2024至今，人工智能与数据科学学院，助理教授

学术发表：

（*共同一作，#通讯作者）

2026:

[1] Zican Dong,Peiyu Liu, Junyi Li, Zhipeng Chen, Han Peng, Shuo Wang, Wayne Xin Zhao#: ForesightKV: Optimizing KV Cache Eviction for Reasoning Models by Learning Long-Term Contribution. International Conference on Machine Learning (ICML), 2026.（CCF-A类）

[2] Yihan Wang,Peiyu Liu#, Runyu Chen, Wei Xu: Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQL. International Conference on Machine Learning (ICML), 2026.（CCF-A类）

[3] Gen Li,Peiyu Liu#: FastV-RAG: Towards Fast and Fine-Grained Video QA with Retrieval-Augmented Generation. Association for Computational Linguistics (ACL), 2026.（CCF-A类）

[4] Yongxuan Wu, Runyu Chen,Peiyu Liu#, Hongjin Qian: LiveLongBench: Tackling Long-Context Understanding for Spoken Texts from Live Streams. Association for Computational Linguistics (ACL) Findings, 2026.

[5] Yihan Wang,Peiyu Liu#, Runyu Chen, Jiaxing Pu, Wei Xu: Squrve: A Unified and Modular Framework for Complex Real-World Text-to-SQL Tasks. Association for Computational Linguistics (ACL) Demo, 2026.

[6] Yihan Wang, Yunjun Chen, Peiyu Liu#, Huiyan Shu, Runyu Chen, Wei Xu: Enhanced Financial Text-to-SQL Generation via Fine-Grained SQL Refinement. Expert Systems With Applications (ESWA) （贸大B类，CCF-C类）

2025:

[1] Zican Dong, Han Peng,Peiyu Liu#, Wayne Xin Zhao#, Dong Wu, Feng Xiao, Zhifeng Wang: Domain-specific Pruning of Large Mixture-of-Experts Models with Few-shot Demonstrations. Neural Information Processing Systems (NeurIPS), 2025.（CCF-A类）

[2]Peiyu Liu, Tianwen Wei, Bo Zhu, Wayne Xin Zhao, Shuicheng Yan: Masks Can be Learned as an Alternative to Experts. Association for Computational Linguistics (ACL), 2025.（CCF-A类）

[3] Jiangbo Pei,Peiyu Liu#, Wayne Xin Zhao#, Aidong Men, Yang Liu: Socratic Style Chain-of-Thoughts Help LLMs to be a Better Reasoner. Association for Computational Linguistics (ACL) Findings, 2025.

[4] Yihan Wang*,Peiyu Liu*, Xin Yang: LinkAlign: Scalable Schema Linking for Real-world Large-scale Multi-database Text-to-SQL. Empirical Methods in Natural Language Processing (EMNLP), 2025.（CCF-B类）

2024:

[1]Peiyu Liu, Ze-Feng Gao, Xin Zhao, Yipeng Ma, Tao Wang, Ji-Rong Wen: Unlocking Data-free Low-bit Quantization with Matrix Decomposition for KV Cache Compression. Association for Computational Linguistics (ACL), 2024.（CCF-A类）

[2]Peiyu Liu, Zikang Liu, Ze-Feng Gao, Dawei Gao, Wayne Xin Zhao, Yaliang Li, Bolin Ding, Ji-Rong Wen: Do Emergent Abilities Exist in Quantized Large Language Models: An Empirical Study. Computational Linguistics (COLING), 2024.（CCF-B类）

[3]Peiyu Liu*, Ze-Feng Gao*, Xiao Zhang, Wayne Xin Zhao, Ji-Rong Wen: Enhancing Parameter-efficient Fine-tuning with Simple Calibration Based on Stable Rank. Computational Linguistics (COLING), 2024.（CCF-B类）

研究项目：

主持国家自然科学基金青年项目“大语言模型的参数-缓存联合优化与自适应压缩研究”，2026年1月-2028年12月。

教授课程：

本科生：《数据结构与算法分析》，《人工智能素养与实践》，《C++算法设计》

研究生：《人工智能前沿技术》，“AI+X”系列课程