• Publications
  • Influence
All That’s ‘Human’ Is Not Gold: Evaluating Human Evaluation of Generated Text
The role untrained human evaluations play in NLG evaluation is examined and three approaches for quickly training evaluators to better identify GPT3-authored text are explored and it is found that while evaluation accuracy improved up to 55%, it did not significantly improve across the three domains.
IUCL at SemEval-2016 Task 6: An Ensemble Model for Stance Detection in Twitter
This work presents the IUCL system, based on supervised learning, for the shared task on stance detection, and shows that random forest is good at retrieving minority classes and gradient boosting majority classes.
Multimodal Alignment for Affective Content
Humans routinely extract important information from images and videos, relying on their gaze. In contrast, computational systems still have difficulty annotating important visual information in a
SpecTran : Neural Network Machine Translator of Javadoc Tags to Java Specifications
This work investigates whether recent advances in neural machine translation systems in the natural language processing community can be easily adapted to this domain of assisting programmers with creating program specifications and test the system on the Javadoc informal specification scheme.