Xingyu Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xingyu Liu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
Jiehui Zhou
1 shared publication- Jiehui Zhou
Alibaba Group, Hangzhou, China
1 shared publication Juntian Zhang
1 shared publicationKam-Kwai Wong
1 shared publicationL. Luan
College of Pharmaceutical Sciences, Zhejiang University, Yuhangtang Road 866, Hangzhou 310058, China
1 shared publication- Liang Sun
DAMO Academy, Alibaba Group, Bellevue, WA, USA
1 shared publication
Latest Publications
- 9 August 2025
- Published by Elsevier in Computers & Graphics
- Vol. 131
CausalPrism: A visual analytics approach for subgroup-based causal heterogeneity exploration
- 9 August 2025
- Published by Elsevier in Computers & Graphics
- Vol. 131
CausalPrism: A visual analytics approach for subgroup-based causal heterogeneity exploration
- Open Access
- 24 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 4619-4630
CURLS: Causal Rule Learning for Subgroups with Significant Treatment Effect
In causal inference, estimating heterogeneous treatment effects (HTE) is critical for identifying how different subgroups respond to interventions, with broad applications in fields such as precision medicine and personalized advertising. Although HTE estimation methods aim to improve accuracy, how to provide explicit subgroup descriptions remains unclear, hindering data interpretation and strategic intervention management. In this paper, we propose CURLS, a novel rule learning method leveraging HTE, which can effectively describe subgroups with significant treatment effects. Specifically, we frame causal rule learning as a discrete optimization problem, finely balancing treatment effect with variance and considering the rule interpretability. We design an iterative procedure based on the minorize-maximization algorithm and solve a submodular lower bound as an approximation for the original. Quantitative experiments and qualitative case studies verify that compared with state-of-the-art methods, CURLS can find subgroups where the estimated and true effects are 16.1% and 13.8% higher and the variance is 12.0% smaller, while maintaining similar or better estimation accuracy and rule interpretability. Code is available at https://osf.io/zwp2k/.
- 24 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 4619-4630
Open AccessCURLS: Causal Rule Learning for Subgroups with Significant Treatment Effect
In causal inference, estimating heterogeneous treatment effects (HTE) is critical for identifying how different subgroups respond to interventions, with broad applications in fields such as precision medicine and personalized advertising. Although HTE estimation methods aim to improve accuracy, how to provide explicit subgroup descriptions remains unclear, hindering data interpretation and strategic intervention management. In this paper, we propose CURLS, a novel rule learning method leveraging HTE, which can effectively describe subgroups with significant treatment effects. Specifically, we frame causal rule learning as a discrete optimization problem, finely balancing treatment effect with variance and considering the rule interpretability. We design an iterative procedure based on the minorize-maximization algorithm and solve a submodular lower bound as an approximation for the original. Quantitative experiments and qualitative case studies verify that compared with state-of-the-art methods, CURLS can find subgroups where the estimated and true effects are 16.1% and 13.8% higher and the variance is 12.0% smaller, while maintaining similar or better estimation accuracy and rule interpretability. Code is available at https://osf.io/zwp2k/.
- 1 January 2014
- Published by Brill in World Mycotoxin Journal
- Vol. 7 (2) , 121-129
Simultaneous determination of aflatoxin B1, aflatoxin B2, mycophenolic acid and sterigmatocystin in grape pomace by UHPLC-MS/MS
A reliable ultrahigh performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) method was developed for the simultaneous determination of aflatoxin B1, aflatoxin B2, mycophenolic acid and sterigmatocystin in grape pomace. The samples were extracted by acetonitrile aqueous solution and further purified using a solid-phase extraction-based homemade clean-up cartridge. Next, the analytes were separated on a reversed-phase C18 column with a mobile phase consisting of water and acetonitrile. The separated compounds were detected with a tandem quadrupole mass spectrometer operating in positive electro-spray ionisation mode using multiple reaction monitoring. The established method was extensively validated by determining linearity (R2≯0.999), recovery (97.5-102.8%) and precision (relative standard deviation ≤7.0%). This method was then used for the simultaneous determination of the four mycotoxins in grape pomace samples.
- 1 January 2014
- Published by Brill in World Mycotoxin Journal
- Vol. 7 (2) , 121-129
Simultaneous determination of aflatoxin B1, aflatoxin B2, mycophenolic acid and sterigmatocystin in grape pomace by UHPLC-MS/MS
A reliable ultrahigh performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) method was developed for the simultaneous determination of aflatoxin B1, aflatoxin B2, mycophenolic acid and sterigmatocystin in grape pomace. The samples were extracted by acetonitrile aqueous solution and further purified using a solid-phase extraction-based homemade clean-up cartridge. Next, the analytes were separated on a reversed-phase C18 column with a mobile phase consisting of water and acetonitrile. The separated compounds were detected with a tandem quadrupole mass spectrometer operating in positive electro-spray ionisation mode using multiple reaction monitoring. The established method was extensively validated by determining linearity (R2≯0.999), recovery (97.5-102.8%) and precision (relative standard deviation ≤7.0%). This method was then used for the simultaneous determination of the four mycotoxins in grape pomace samples.