Marcella Cornia

Papers

Total Citations

H-Index

About

Marcella Cornia is a leading researcher at the intersection of computer vision, robotics, and human-robot interaction, with a primary focus on developing embodied agents that can intelligently explore environments and communicate with humans. Her work centers on creating autonomous systems that combine deep reinforcement learning with natural language generation to enable robots to not only navigate complex indoor spaces but also explain their actions and perceptions. Cornia’s most impactful contribution is her 2022 paper “Focus on Impact: Indoor Exploration With Intrinsic Motivation” (19 citations), which introduced hierarchical deep neural agents trained with DRL for efficient exploration. She has further advanced the field through her work on “Embodied Agents for Efficient Exploration and Smart Scene Description” (5 citations), pushing toward seamless human-robot communication. A notable achievement is her 2020 paper “SMArT: Training Shallow Memory-aware Transformers for Robotic Explainability” (3 citations), which pioneered explainable AI in robotics by generating natural language explanations from visual data. Her most recent work (2025) on modeling human gaze behavior with diffusion models demonstrates her continued innovation in understanding visual attention. With a growing citation impact, Cornia is establishing herself as a key voice in creating transparent, communicative robotic systems that can operate alongside humans in real-world environments.