Talks and Presentations

Does Robin Hood Use a Lightsaber?: Automated Planning for Storytelling

February 01, 2024

Conference proceedings talk, AAAI/SIGAI Doctoral Consortium, Association for the Advancement of Artificial Intelligence (AAAI), Vancouver, BC, Canada

Automated Planning can therefore be combined with Natural Language text generation to create narratives (stories) that are logical, coherent, and believable. A planning model provides scaffolding to an LLM so that the LLM’s language generation is context-dependent, in order to allow users to create more coherent, logical, and believable stories in a variety of domains.

TattleTale: Storytelling with Planning and Large Language Models

June 15, 2022

Conference proceedings talk, The International Conference on Automated Planning and Scheduling (ICAPS) SPARK workshop, Virtual

We demonstrate the use of a planning model that provides scaffolding to an LLM so that its language generation is context dependent in order to create more coherent and believable stories in a variety of domains.

A Natural Language Model for Generating PDDL

August 05, 2021

Conference proceedings talk, The International Conference on Automated Planning and Scheduling (ICAPS) KEPS workshop, Virtual

The goal of this preliminary work is to predict the next completion in PDDL code, based on previous and surrounding text. Generating valid PDDL code is a key component in creating robust planners. Thus, the ability to generate PDDL code will be extremely useful to PDDL practitioners for the purpose of solving planning problems. It further opens the door to providing a source of inspiration for the modeller. The main contribution of our approach is a language model built using Recurrent Neural Networks (RNNs) that is trained on existing PDDL domains, which can be used to generate PDDL-like code.

Automatic Term Extraction in Technical Domain using Part-of-Speech and Common Word Features

August 28, 2018

Conference proceedings talk, Association for Computing Machinery (ACM) Symposium DocEng 2018, Halifax, NS

Extracting key terms from technical documents allows us to write effective documentation that is specific and clear, with minimum ambiguity and confusion caused by nearly synonymous but different terms. For instance, in order to avoid confusion, the same object should not be referred to by two different names (e.g. “hydraulic oil filter”). In the modern world of commerce, clear terminology is the hallmark of successful RFPs (Requests for Proposal) and is therefore a key to the growth of competitive organizations. While Automatic Term Extraction (ATE) is a well-developed area of study, its applications in the technical domain have been sparse and constrained to certain narrow areas such as the biomedical research domain. We present a method for Automatic Term Extraction (ATE) for the technical domain based on the use of part-of-speech features and common words information.