Kais Dukes

Papers

Total Citations

H-Index

About

Kais Dukes is a researcher whose work sits at the intersection of natural language processing (NLP), semantic parsing, and robotics. His primary research focus has been on developing systems that allow robots to understand and execute commands given in natural language, a critical step toward more intuitive human-robot interaction. Dukes’s most significant contribution is his leadership in creating the “SemEval-2014 Task 6: Supervised Semantic Parsing of Robotic Spatial Commands,” which provided a high-quality, annotated dataset that became a benchmark for the field. This work, his most cited with 25 citations, advanced the challenge of contextual parsing, where spatial scene information helps disambiguate language. He further extended this line of inquiry with the “Robot Commands Treebank,” a crowdsourced resource for contextual parsing, and the “Train Robots” dataset, which uses synthetic scenes of a robotic arm to generate natural language descriptions via Amazon Mechanical Turk. By combining crowdsourcing with structured semantic representations, Dukes has helped lay the groundwork for more robust, context-aware robotic command understanding, directly impacting how researchers approach grounded language learning in embodied agents.