Max Robotics — Global Service Robot Directory + US Certification

About

Yingli Zhao is a rising researcher at the intersection of embodied AI and multimodal machine learning, whose work bridges the gap between intelligent systems and dynamic, human-centered environments. Her most impactful contribution, "Crowd-Comfort Robot Navigation Among Dynamic Environment Based on Social-Stressed Deep Reinforcement Learning" (2021), has garnered 16 citations and introduces a novel framework for socially-aware robot movement. By modeling pedestrian stress and comfort as a core reward signal, Zhao’s algorithm enables robots to navigate crowded spaces with unprecedented sensitivity to human social norms—a critical advance for service robots in hospitals, airports, and public squares. More recently, Zhao has tackled one of AI’s grand challenges: unified multimodal learning. Her 2026 paper, "Multimodal learning with next-token prediction for large multimodal models," proposes extending the next-token prediction paradigm—the engine behind large language models—to simultaneously process text, images, and video. This work promises to simplify multimodal architectures while preserving generative power. Though early in its citation life, this contribution signals Zhao’s ambition to reshape how AI systems perceive and generate across modalities. Her trajectory—from crowd-aware navigation to foundational multimodal learning—marks her as a versatile thinker driving both practical robotics and theoretical AI forward.