Shikun Liu

Peking University

Papers

2

Total Citations

22

H-Index

2

About

Shikun Liu is a researcher whose work bridges artificial intelligence and computational methods, with a primary focus on multi-task learning and its applications in complex domains. His most significant contribution, "Auto-Lambda: Disentangling Dynamic Task Relationships" (2022, 18 citations), tackles a fundamental challenge in multi-task learning: the prohibitive computational cost of modeling pairwise task interactions. Liu introduced a novel framework that dynamically disentangles task relationships without exhaustive pairwise training, enabling more efficient and scalable multi-task models that improve generalization across related tasks. This work has been recognized for its potential to advance AI systems that must handle multiple objectives simultaneously. More recently, Liu has expanded into interdisciplinary applications, co-authoring "The evolution and integration of technology in spinal neurosurgery: A scoping review" (2024, 4 citations), demonstrating his ability to apply computational thinking to critical healthcare challenges. His research is particularly relevant for students and researchers interested in efficient multi-task learning, task relationship modeling, and the intersection of AI with medical technology.

Research Focus

Key Achievements

2
H-Index
2
Papers
22
Total Citations
11
Avg Citations/Paper
🏆 Most Cited Paper
Auto-Lambda: Disentangling Dynamic Task Relationships
18 citations · 2022
📈 Most Prolific Year: 2022 (1 Papers)
🤝 Key Collaborators: 9
🏛 Institutions: Peking University

Top Papers

  1. 1
  2. 2

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago