Papers
5
Total Citations
92
H-Index
5
About
Song Tang is a leading researcher in cognitive robotics, specializing in multimodal human-robot interaction, sensory adaptation, and natural language grounding. Their work bridges the gap between human-like perception and robotic intelligence, with a focus on enabling robots to understand and adapt to dynamic environments. Tang’s most cited paper (30 citations) introduces object affordance-based multimodal fusion for natural interaction, while their 2021 work on model adaptation via hypothesis transfer with gradual knowledge distillation (19 citations) addresses unsupervised domain adaptation for changing environments. A standout contribution is their bioinspired memristor-based sensory memory system (18 citations), which mimics human sensory adaptation to allow robots to gradually adjust sensitivity based on recent stimuli. Tang has also advanced natural language grounding through intention-related semantic extraction (13 citations) and interactive scene graph parsing (12 citations), enabling robots to interpret ambiguous human instructions. Their interdisciplinary approach—combining psychology, neuroscience, and robotics—has positioned them as a pioneer in creating more intuitive, adaptive robotic systems. With a growing citation impact and innovative hardware-software solutions, Tang’s work is shaping the future of intelligent, human-aware robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5