Profiles2021-2026
Zhi Yang
Peking University
15Papers
1Last author
25Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers15
- TileLang: Bridge Programmability and Performance in Modern Neural KernelsAuthor 12 of 12ICLR 2026
- Towards All-Atom Foundation Models for Biomolecular Binding Affinity PredictionAuthor 5 of 6ICLR 2026
- SDPipe: A Semi-Decentralized Framework for Heterogeneity-aware Pipeline-parallel TrainingAuthor 3 of 4VLDB 2023 · 48 citations
- HET: Scaling out Huge Embedding Model Training via Cache-enabled Distributed FrameworkAuthor 5 of 7VLDB 2022 · 70 citations
- PaSca: A Graph Neural Architecture Search System under the Scalable ParadigmAuthor 8 of 9WWW 2022 · 69 citations
- Model Degradation Hinders Deep Graph Neural NetworksAuthor 7 of 8KDD 2022 · 42 citations
- TSPLIT: Fine-grained GPU Memory Management for Efficient DNN Training via Tensor SplittingAuthor 3 of 4ICDE 2022 · 26 citations
- HET-GMP: A Graph-based System Approach to Scaling Large Embedding Model TrainingAuthor 6 of 7SIGMOD 2022 · 24 citations
- Information Gain Propagation: a New Way to Graph Active Learning with Soft LabelsAuthor 7 of 8ICLR 2022 · 24 citations
- Deep and Flexible Graph Neural Architecture SearchAuthor 5 of 6ICML 2022 · 5 citations
- Node Dependent Local Smoothing for Scalable Graph LearningAuthor 7 of 8NeurIPS 2021 · 87 citations
- Heterogeneity-Aware Distributed Machine Learning Training via Partial ReduceAuthor 4 of 6SIGMOD 2021 · 64 citations
- Grain: Improving Data Efficiency of Graph Neural Networks via Diversified Influence MaximizationAuthor 2 of 7VLDB 2021 · 60 citations
- RIM: Reliable Influence-based Active Learning on GraphsAuthor 7 of 8NeurIPS 2021 · 43 citations
- ROD: Reception-aware Online Distillation for Sparse GraphsAuthor 8 of 9KDD 2021 · 22 citations
