Alex Lascarides

About

Alex Lascarides is a researcher whose work sits at the intersection of human-robot interaction, natural language understanding, and machine learning. Her research focuses on enabling robots to interpret and act upon human instructions by grounding abstract linguistic concepts in high-dimensional sensory data — a critical challenge for deploying robots in real-world, semi-structured environments. Among her most significant contributions is her work on learning from demonstration, where she has explored how robots can efficiently acquire new behaviors from human users with minimal data. Her papers on interpretable latent spaces (2018, 15 citations) and disentangled relational representations (2019, 10 citations) advance this field by introducing structured, interpretable frameworks that improve both learning efficiency and explainability — moving beyond opaque deep learning models toward systems that can reason transparently. More recently, Lascarides has extended her research into large language models, co-authoring work on dialogue-based generation of autonomous driving simulation scenarios (2023, 9 citations), reflecting her interest in leveraging modern generative AI for safety-critical applications. Across her career, her research consistently bridges computational linguistics and robotics, making her a distinctive voice in grounded language understanding and human-centered AI systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Interpretable Latent Spaces for Learning from Demonstration

   15 citations · 2018
2. 2
   Disentangled Relational Representations for Explaining and Learning from\n Demonstration

   10 citations · 2019
3. 3
   Dialogue-based generation of self-driving simulation scenarios using Large Language Models

   9 citations · 2023
4. 4
   6 citations

Key Collaborators

Contact & Links