学科专业(学术型/专业型): 计算机应用技术,计算机技术
人工智能、大模型、智能体、智算
王浩,研究员、博士生导师,院领军人才。长期致力于通用人工智能(AGI)及 AI 基础设施前沿研究,旨在赋予机器类人的思考、规划与自主决策能力,以技术创新驱动产业智能化升级。
深耕学术界与工业界多年,曾先后任职于中国科学院软件研究所、奇虎 360 搜索事业部、阿里巴巴达摩院及阿里云智能集团,具备百人规模产品与研发团队的管理经验。学术领域,在 KDD、NeurIPS、SIGIR、WWW、CVPR、ICCV、ECCV、IJCAI、AAAI、ICDE、ACL、TPAMI、TOIS、TKDE 等数据科学与人工智能领域顶级会议及期刊发表论文百余篇;并多次主持国家自然科学基金(面上/青年)、教育部留学回国人员科研启动基金及北京市自然科学基金等项目,实现了前沿科研与产业落地的深度融合。
谷歌学术:https://scholar.google.com/citations?user=N4rQYDAAAAAJ&hl=en
1、Li, W., Guo, J., Wang, Y., Xiao, H., Zhang, Y., Liu, G., & Wang, H. (2026). Evo-Retriever: LLM-Guided Curriculum Evolution with Viewpoint-Pathway Collaboration for Multimodal Document Retrieval. CVPR 2026
2、Zhao, Y., Zhu, H., Jiang, T., Li, S., Xu, X., & Wang, H. (2026). Co-EPG: A Framework for Co-Evolution of Planning and Grounding in Autonomous GUI Agents. AAAI 2026
3、Quan, G., Feng, W., Hao, C., Jiang, G., Zhang, Y., & Wang, H.(2025). RASD: Retrieval-augmented speculative decoding. ACL 2025
4、Huang, K., Zou, H., Wang, B., Xi, Y., Xie, Z., & Wang, H. (2025). AirCache: Activating Inter-modal Relevancy KV Cache Compression for Efficient Large Vision-Language Model Inference. ICCV 2025
5、Zhang, C., Wang, H., Jiang, F., & Yin, H. (2021). Adapting to context-aware knowledge in natural conversation for multi-turn response selection. WWW 2021
6、Chen, T., Yin, H., Ren, J., Huang, Z., Zhang, X., & Wang, H. (2021). Uniting heterogeneity, inductiveness, and efficiency for graph representation learning. IEEE Transactions on Knowledge and Data Engineering (TKDE), 35(2), 2103-2117.
7、Ren, X., Yin, H., Chen, T., Wang, H., et.al (2020). CRSAL: Conversational Recommender Systems with Adversarial Learning. ACM Transactions on Information Systems (TOIS), 38(4), 1-40.
8、Wang, H., Fu, Y., Wang, Q., Yin, H., Du, C., Xiong, H. (2017). A Location-Sentiment-Aware Recommender System for Both Home-Town and Out-of-Town Users. KDD 2017
9、Wang, H., Wang, W., Zhang, C., & Xu, F. (2014). Cross-domain metric learning based on information theory. AAAI 2014
10、Wang, H., & Ohsawa, Y. (2013). Idea discovery: A scenario-based systematic approach for decision making in market innovation. Expert Systems with Applications, 40(2), 429-438.