Welcome to XLearning Lab
XLearning Lab is affiliated with the School of Computer Science at Sichuan University and the National Key Laboratory for Fundamental Algorithms and Models for Engineering Simulation. Situated on the historic Wangjiang Campus, the lab was founded and is currently led by Prof. Xi Peng.
The name XLearning stands for "Learning for X and Learning with X." While the former reflects our commitment to interdisciplinary research, the latter highlights the innate curiosity of the XLearning community. Currently, our research focuses on unsupervised/self-supervised learning, multi-modal learning, AI4LifeScience, and Physical AI.
Recent Works
ICML
Multiview Self-Representation Learning across Heterogeneous Views
Jie Chen, Zhu Wang, Chuanbin Liu, Xi Peng
ICML (Oral, acc rate=0.7%)
DOUBT: Decoupled Object-level Understanding and Bridging via vMF-based Trustworthiness for Hallucination Detection in MLLMs
Kaiqi Chen, Yang Qin, Changhao He, Xi Peng, Peng Hu
View PaperICML
RLSF-V: Mitigating Hallucinations in MLLMs via Fuzzy Semantic Self-Feedback
Changhao He, Shuhaoyan, Shuxian Li, Xi Peng, Peng Hu
View PaperICML
Reliable Thinking with Images
Haobin Li, Yutong Yang, Yijie Lin, Daixiang, Mouxing Yang, Xi Peng
View PaperTPAMI
Deep Information-Balanced Multimodal Learning
Yang Qin, Yanglin Feng, Yuan Sun, Dezhong Peng, Xi Peng, Peng Hu
View PaperTPAMI
Learning with Partial and Noisy Correspondence in Graph Matching
Yijie Lin, Mouxing Yang, Peng Hu, Jiancheng Lv, Hao Chen, Xi Peng
View Paper