计算机科学与技术学院
软件所所长，教授，博士生导师
近年来共承担科研课题30余项，其中国家自然科学基金和863项目7项，省部级科研课题10余项；获陕西省自然科学二等奖、中国人民解放军科技进步奖等省部级科研奖励5项；现为IEEE Member、ACM Member、中国计算机学会高级会员；在相关领域先后发表包括NeurIPS、ACL、AAAI、ACM MM、IJCAI、EMNLP、ICASSP等高水平国际会议，以及IEEE TKDE、IEEE TMM、ACM CSUR、IEEE TAFFC、IEEE TNNLS等国际权威期刊论文200余篇，其中 SCI 70余篇；授权发明专利30余项。
E-mail：yxyphd@mail.xjtu.edu.cn
-------------------------------------
学院：计算机科学与技术学院
单位：西安交通大学
地址：西安 陕西 中国
邮编：710049
本团队每年招收硕士生10-15名，博士生3-5名。主要研究方向：情感计算、音乐人工智能、图形图像。欢迎有计算机、自动化、数学等专业背景的学生加入！
Zhang, Yang Xinyu and X. Bai, "Crop the Way You Like: Personalized Image Cropping by Integrating Subjective and Objective Features," in IEEE Transactions on Multimedia, doi: 10.1109/TMM.2026.3651046.(CCF-A,TMM)
Zhang S, Yang Xinyu. RetouchAgent: Towards Interactive and Explainable Image Retouching with MLLM Agents[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2026, 40(35): 29901-29910.(CCF-A,AAAI)
Chen H, Yang Xinyu, Zhu J, et al. Skill Path: Unveiling Language Skills from Circuit Graphs[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2026, 40(36): 30210-30217.(CCF-A,AAAI)
Chen H, Zhu J, Yang Xinyu, et al. CLUE: Conflict-guided Localization for LLM Unlearning Framework[C]//The Fourteenth International Conference on Learning Representations.(CCF-A,ICLR)
Du K, Kang Y, Yang Xinyu, et al. ST-HHOL: Spatio-Temporal Hierarchical Hypergraph Online Learning for Crime Prediction[C]//The Fourteenth International Conference on Learning Representations.(CCF-A,ICLR)
Du K, Yang Xinyu, Chen H. CaASR: A Causal Lens for Refining Temporal Action Segmentation[J]. IEEE Transactions on Multimedia, 2026.(CCF-A,TMM)
Li C, Yang Xinyu, Yang W, et al. VaF-LangSplat: Voxel-Aware Fusion Language Gaussian Splatting[C]//Proceedings of the 33rd ACM International Conference on Multimedia. 2025: 4952-4961.(CCF-A,MM)
Chen H, Yang Xinyu, Zhu J, et al. Quantifying semantic emergence in language models[C]//Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025: 12041-12054.(CCF-A,ACL)
Chen H, Zhu J, Yang Xinyu, et al. Rethinking Circuit Completeness in Language Models: AND, OR, and ADDER Gates[C]//The Thirty-ninth Annual Conference on Neural Information Processing Systems.(CCF-A,NeurIPS)
Zhao D, Yang Xinyu, Chen H. Debiasing the Fine-Grained Classification Task in LLMs with Bias-Aware PEFT[C]//Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025: 14731-14746.(CCF-A,ACL)
Hu G, Kollias D, Yang Xinyu. Grounding Emotion Recognition with Visual Prototypes: VEGA-Revisiting CLIP in MERC[C]//Proceedings of the 33rd ACM International Conference on Multimedia. 2025: 5667-5676.(CCF-A,MM)
Mu Z, Yang Xinyu, Wang G. SepALM: Audio Language Models Are Error Correctors for Robust Speech Separation[C]//Proceedings of the 34th International Joint Conference on Artificial Intelligence, IJCAI 2025. International Joint Conferences on Artificial Intelligence, 2025: 8204-8212.(CCF-A,IJCAI)
Mu Z, Chen R, Li A, et al. From Continuous to Discrete: Cross-Domain Collaborative General Speech Enhancement via Hierarchical Language Models[C]//Proceedings of the 33rd ACM International Conference on Multimedia. 2025: 219-228.(CCF-A,MM)
