Saved in:
Bibliographic Details
Main Authors: Delafuente, Patricia, Honraopatil, Arya, Martin, Lara J.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.18112
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918164627980288
author Delafuente, Patricia
Honraopatil, Arya
Martin, Lara J.
author_facet Delafuente, Patricia
Honraopatil, Arya
Martin, Lara J.
contents This paper explores the application of Large Language Models (LLMs) and reasoning to predict Dungeons & Dragons (DnD) player actions and format them as Avrae Discord bot commands. Using the FIREBALL dataset, we evaluated a reasoning model, DeepSeek-R1-Distill-LLaMA-8B, and an instruct model, LLaMA-3.1-8B-Instruct, for command generation. Our findings highlight the importance of providing specific instructions to models, that even single sentence changes in prompts can greatly affect the output of models, and that instruct models are sufficient for this task compared to reasoning models.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18112
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Does Reasoning Help LLM Agents Play Dungeons and Dragons? A Prompt Engineering Experiment
Delafuente, Patricia
Honraopatil, Arya
Martin, Lara J.
Computation and Language
This paper explores the application of Large Language Models (LLMs) and reasoning to predict Dungeons & Dragons (DnD) player actions and format them as Avrae Discord bot commands. Using the FIREBALL dataset, we evaluated a reasoning model, DeepSeek-R1-Distill-LLaMA-8B, and an instruct model, LLaMA-3.1-8B-Instruct, for command generation. Our findings highlight the importance of providing specific instructions to models, that even single sentence changes in prompts can greatly affect the output of models, and that instruct models are sufficient for this task compared to reasoning models.
title Does Reasoning Help LLM Agents Play Dungeons and Dragons? A Prompt Engineering Experiment
topic Computation and Language
url https://arxiv.org/abs/2510.18112