This repository is a copy of https://github.com/philotron/conversational-grounding-llm
It contains the source code and data for the study conducted at AIRC in November 2023. The work is presented in the following paper, published at IWSDS-2024:
Kristiina Jokinen, Phillip Schneider, Taiga Mori (2024). Towards Harnessing Large Language Models for Enhanced Comprehension of Conversational Grounding. Proceedings of the 14th International Workshop on Spoken Dialogue Systems Technology (IWSDS 2024), 4-6 March 2024, Sapporo, Japan.
-
Notebook: The Jupyter notebook (
conv-grounding-llm-prediction.ipynb
) details the experimental procedures, prompts, and configuration of the large language model. -
Datasets: The dataset
annotated_dialogues.xlsx
was utilized in the experiment as it contains the input dialogues. The LLM predictions are provided as four additional datasets.
- Open the Jupyter notebook using a compatible environment or platform.
- Follow the step-by-step instructions to reproduce the experiment.