2024年度第7回定例研究会「Prompt Engineering for Game Content Generation Using Large Language Models」

来る2024年7月25日(木曜日)、立命館大学ゲーム研究センターによる2024年度第7回定例研究会を実施致します。発表者は、Pittawat Taveekitworachai氏です。登録・参加料不要となっております。お誘い合わせの上、奮ってご参加のほど、お待ちしております。

The Ritsumeikan Center for Game Studies will hold its seventh workshop
of the 2024 academic year on Thursday, July 25, 2024. The presenter will
be Mr. Pittawat Taveekitworachai. Registration is not required, and
there is no participation fee. We look forward to your active participation.

■発表タイトル Title
Prompt Engineering for Game Content Generation Using Large Language Models

■発表者 Presenter
Pittawat Taveekitworachai (GISE, Ritsumeikan University)
https://scholar.google.com/citations?user=2CFoP9cAAAAJ&hl=en

■日時 Date and Time
7月25日(木曜日) 17:00~18:30
July 25 (Thu) 17:00-18:30

■場所 Venue
立命館大学 大阪茨木キャンパス H棟 H702号室 アクセス
(https://www.ritsumei.ac.jp/accessmap/oic/)
Ritsumeikan Osaka Ibaraki Campus, Building H, Room H702 Access
(https://www.ritsumei.ac.jp/accessmap/oic/)

■概要 Summary
Large language models (LLMs) are remarkable and versatile across various
domains of applications. However, their performance can vary greatly
depending on how we interact with them. In this talk, we will explore
how we can effectively communicate with LLMs, starting from simple
zero-shot prompting to more advanced techniques like chain-of-thought
prompting. There is an opportunity to implement more complex prompt
engineering approaches, such as tree-of-thought prompting, which
requires programming and demonstrates that prompt engineering is not
simply the crafting of text. This hands-on session utilizes a game
content generation platform as a case study to demonstrate the
effectiveness of prompt engineering. ChatGPT4PCG is a competition aimed
at finding ways to effectively ask LLMs to generate a level of an
instructed target structure. Using the newest resources from the
competition, you will also have a chance to set up and run LLMs locally
on your computer! Finally, we will touch on a broader perspective and
breadth of prompt engineering, and offer some tips on conducting
research involving prompt engineering.

■このイベントに参加する予定がある場合は、次のページをご覧ください。

If you plan to attend this event, please visit the following page.
https://chatgpt4pcg.github.io/tutorial