Papers
3
Total Citations
23
H-Index
3
About
Mengxiao Hu is a rising researcher at the intersection of artificial intelligence, robotics, and cybersecurity, with a focus on enabling robots to learn complex skills from human demonstrations. Their work addresses two critical challenges: securing autonomous multi-robot systems and overcoming the data bottleneck in robotic imitation learning. Hu’s most cited paper, “Secure and smart autonomous multi-robot systems for opinion spammer detection” (2021, 16 citations), pioneers the use of multi-robot teams for cybersecurity tasks, demonstrating how collaborative AI can detect deceptive online behavior. More recently, Hu has advanced visual imitation learning with “GraphMimic” (2025, 4 citations), which introduces a graph-to-graphs generative model that extracts policy knowledge from videos, dramatically reducing the need for expensive action-labeled robot data. Building on this, “FMimic” (2025, 3 citations) leverages foundation models—specifically vision language models—to enable fine-grained action learning from human videos, pushing the boundaries of how robots can acquire new skills through observation. Hu’s contributions are particularly notable for bridging the gap between data-efficient learning and real-world deployment, with potential applications in manufacturing, healthcare, and autonomous systems. Their work is already shaping how researchers think about scalable, secure robotic intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3