Minori Toyoda

Papers

Total Citations

H-Index

About

Minori Toyoda is a pioneering researcher at the intersection of robotics and natural language processing, whose work focuses on grounding abstract linguistic concepts in physical robot actions. Her key research areas include embodied cognition, cross-modal learning, and human-robot interaction, where she addresses the fundamental challenge of enabling robots to understand and generate language through direct physical experience. Toyoda's major contribution lies in developing neural network models that bridge the gap between symbolic language and continuous motor control. Her most cited work, "Embodying Pre-Trained Word Embeddings Through Robot Actions" (2021, 19 citations), introduces a groundbreaking approach for robots to acquire grounded representations of actions and their linguistic descriptions, including the ability to handle polysemous words—a critical capability for natural human-robot collaboration. She further advanced this field with "Learning Bidirectional Translation Between Descriptions and Actions With Small Paired Data" (2022, 7 citations), which demonstrates that robots can learn to translate between language and actions using minimal paired training data, overcoming a major bottleneck in robotics research. Her work has significant implications for developing robots that can seamlessly integrate into human environments, understand nuanced instructions, and communicate their own actions effectively. Toyoda's research represents a crucial step toward truly intelligent, language-capable robots.