Papers
1
Total Citations
2
H-Index
1
About
Tianyu Wu is a rising researcher at the forefront of robotics and artificial intelligence, whose work bridges sequence modeling and lifelong learning for autonomous systems. His primary research areas include robotic manipulation, recurrent neural architectures, and decision-making frameworks for embodied AI. Wu’s most notable contribution is the development of Decision-RWKV, a novel recurrent sequence modeling approach that addresses the critical challenge of lifelong learning in robotic manipulation. By reimagining transformer-based architectures for continuous adaptation, his work enables robots to acquire and retain manipulation skills over extended periods without catastrophic forgetting. Although his seminal 2024 paper has garnered 2 citations in its early stages, the innovative fusion of RWKV (Receptance Weighted Key Value) models with decision-making processes represents a significant departure from conventional attention-based methods, offering computational efficiency and scalability for real-world robotics. Wu’s research stands at the intersection of NLP-inspired architectures and physical action, promising to reshape how robots learn and adapt in dynamic environments. As the field increasingly demands energy-efficient, continuously learning systems, his work positions him as a forward-thinking contributor to next-generation embodied intelligence.
Research Focus
Key Achievements
Top Papers
- 1