


A session held on August 23rd, the third day of CEDEC 2024 ``Almost a full AI game. A mechanism that allows Red Ram to generate an unlimited number of mystery games of the player's preference'' .
"Red Ram" is a technology demo developed by Morikatron, a developer specializing in game AI, and is an attempt to automatically generate a murder mystery-style culprit guessing game based on keywords entered by the player. It has become.
In this session, Shigenori Miyamoto, Morikatron's lead engineer who was in charge of natural language processing for the game, and Rikito Takahashi, an AI engineer who was in charge of system design and client implementation, explained how the generation works. It was revealed.
Lead engineer Shigenori Miyamoto (right) and AI engineer Rikito Takahashi (left)
![]() |
The session started with Mr. Takahashi talking about the message he had put into this session. That is, ``I want more games to use generation AI in-game.'' This is a message unique to Morikatron, which is skilled in AI technology, but the company believes that by utilizing generative AI in games, two types of fun can be created.
One is that there are so many variations that you won't get bored. Another benefit is the ability to create a gaming experience tailored to each player's preferences. ``Red Ram'', which will be discussed in this session, was created to study the gaming experience unique to generative AI.
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Encik Miyamoto menerangkan butiran generasi ini. Pertama, plot dijana secara berperingkat menggunakan LLM (model bahasa berskala besar/``Red Ram'' menggunakan GPT 3.5 dan 4 OpenAI) berdasarkan kata kunci yang dimasukkan oleh pemain. Jika keadaan kompleks diperlukan untuk penjanaan, GPT 4 berprestasi tinggi nampaknya digunakan.
Sebagai contoh, walaupun pemain memasukkan sesuatu yang tidak sesuai sebagai senjata, LLM akan mencipta senario yang sewajarnya.
![]() |
Dalam penjanaan plot, bukan sahaja latar kejadian tetapi juga panduan penulisan misteri dan cara mencipta aliran kemajuan dimasukkan sebagai gesaan. Sediakan berbilang aliran kemajuan yang dijana secara manual oleh LLM terlebih dahulu dan gunakannya secara rawak. Dikatakan bahawa penjanaan plot tanpa sebarang kekangan format boleh dijangka menghasilkan produk yang lebih berkualiti.
![]() |
Daripada plot yang dibuat dengan cara ini, data cerita dijana dalam format yang boleh dikendalikan oleh program. Seperti yang anda boleh lihat dalam slaid di bawah, pada peringkat plot ia hanyalah teks biasa, tetapi data yang diperlukan seperti lokasi dan mangsa diekstrak daripadanya, dan kandungannya disempurnakan pada masa yang sama.
![]() |
Seterusnya, data terperinci untuk setiap orang, lokasi dan bukti yang muncul dalam permainan dijana daripada data cerita. Selain menentukan profil seseorang seperti umur, personaliti dan pekerjaan, nada suara mereka juga ditentukan. Pada masa ini, gesaan untuk menjana imej juga dibuat.
![]() |
Perkara terakhir yang dihasilkan ialah senario perbualan -- teks soal siasat (baris perbualan) yang dilakukan oleh pemain, detektif, terhadap suspek. Gesaan dibuat dengan memasukkan data yang dijana setakat ini ke dalam enam jenis templat bergantung pada situasi dan lokasi kejadian, dan perbualan dijana. Selain itu, emosi watak disimpulkan daripada teks dialog kepada LLM dan diklasifikasikan kepada enam jenis emosi. Ini dikatakan digunakan untuk menukar ekspresi muka suspek dalam penjanaan imej.
![]() |
Seterusnya, beralih kepada penjanaan imej. Berdasarkan teks yang dijana oleh LLM, imej disediakan menggunakan "Resapan Stabil" (nama model AI penjanaan imej). Untuk imej latar belakang dan bukti, gunakan gesaan yang dibuat apabila anda menjana butiran masing-masing. Pada masa ini, pelarasan dibuat untuk melaraskan sudut tontonan imej latar belakang dan menggantikan serta melukis item bukti yang sukar diwakili.
![]() |
Begitu juga, potret watak menggunakan gesaan yang dibuat apabila data terperinci dijana. Kekangan dikenakan supaya imej dilukis dalam kedudukan tetap dan pada saiz tetap, dan pemprosesan ketelusan latar belakang (memotong) juga dilakukan pada ketika ini.
Untuk pemangkasan, Segmentasi Semantik (kaedah pembahagian yang melabelkan setiap piksel) dan teknologi yang menganggarkan kedalaman imej digunakan untuk memastikan pemangkasan yang boleh dipercayai dengan mengekstrak bahagian dalam sahaja.
![]() |
![]() |
Cabaran dan prospek masa depan untuk permainan AI generatif
Slaid seterusnya menunjukkan kejadian yang sebenarnya dijana sebagai hasilnya.
![]() |
Pertama sekali, memandangkan senjata pembunuhan itu adalah ``tauhu'', adalah wajar untuk mengatakan bahawa punca kematian adalah sesak nafas. Pentas itu ditetapkan sebagai ``syarikat pembangunan permainan yang inovatif,'' jadi tempat seperti pejabat atau bilik persidangan telah dijana. Memandangkan motif jenayah itu adalah "masalah kewangan"
, hubungan manusia yang tertumpu kepada peminjaman dan peminjaman wang telah diwujudkan. Ngomong-ngomong, nampaknya anda boleh memilih antara gaya aksi langsung dan gaya anime untuk watak.
![]() |
![]() |
In order to achieve a consistent scenario, it is necessary to further devise the system design and the prompts given to the LLM. Other issues such as cost and waiting time were also raised, but these are expected to be gradually resolved through the evolution of the LLM model.
According to Mr. Takahashi, in the future he would like to create an experience where the information entered by the player is reflected in the game world in real time, and in July 2024, he will release a technology demo called "Labyrinth of Kotodama". are.
![]() |
This demo is a text adventure that aims to conquer a dungeon, but the system is such that items are generated based on the words entered by the player. In response to the input words, the item's image, parameters, flavor text, etc. are generated on the spot, and the main character is shown using the item.
![]() |
Additionally, papers and generation examples related to "Red Ram" are available online. It can be viewed using the QR code on the slide below, so if you are interested, why not check it out?
![]() |
The above is the detailed content of What is the mechanism of 'Red Ram' that can generate mystery games by leaving it to AI? [CEDEC 2024]. For more information, please follow other related articles on the PHP Chinese website!

Forbes' NYT Mini Crossword Solutions for Wednesday, April 9th By Paul Tassi The NYT Mini crossword, unlike its larger counterpart, is freely available on the New York Times website or the NYT Games app. However, accessing past puzzles requires an NY

Forbes NYT Strands Solution for Thursday, April 10th: Unraveling the "What's the Buzz?" Mystery Need help conquering today's New York Times Strands puzzle? This guide provides hints and the complete solution for Thursday, April 10th's &qu

Blue Prince: 2025's Top-Scoring Game—A Must-Play Puzzle Adventure With a stunning Metascore of 92, Blue Prince is currently the highest-rated game of 2025, making it a strong contender for Game of the Year. While Grand Theft Auto 6 is anticipated, B

Decrypt the New York Times Connections game: Clues and answers for Thursday, April 10 The weekend is coming, and most people are happy about it, right? However, today I was extremely upset because I fell into a trap in today's "Connections" game! I was blinded by my arrogance. Friends who like some form of entertainment will surely discover this obvious "red herring" soon. But hope today's tips can help you avoid the same traps as me. Before I start, I would like to remind you that I have created a discussion group on Discord for the New York Times Connections game and this column. There is a great little community out there, and we talk about much more than "

Wordle #1391 Hints, Clues, and the Answer for Thursday, April 10th Ready to tackle today's Wordle challenge? This guide provides hints, clues, and the solution for Wordle #1391, played on Thursday, April 10th. Remember, it's 2XP Friday, so your Com

The Esports World Cup returns for its second year, aiming for sustainable growth while offering substantial rewards to attract top teams. This year's prize pool boasts a significant $70 million, a strategic increase from last year's $60 million. Th

The New York Times game Connections is known for its clever wordplay, often employing "traps" or "red herrings"—words fitting multiple categories—to challenge players. This is a key element of the game's appeal, adding a layer of

Teenage audiences are taking movie-going to a new, disruptive extreme. The recent release of a major video game adaptation has seen unprecedented levels of disruptive behavior, primarily involving the excessive and inappropriate use of popcorn as pr


Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

AI Hentai Generator
Generate AI Hentai for free.

Hot Article

Hot Tools

mPDF
mPDF is a PHP library that can generate PDF files from UTF-8 encoded HTML. The original author, Ian Back, wrote mPDF to output PDF files "on the fly" from his website and handle different languages. It is slower than original scripts like HTML2FPDF and produces larger files when using Unicode fonts, but supports CSS styles etc. and has a lot of enhancements. Supports almost all languages, including RTL (Arabic and Hebrew) and CJK (Chinese, Japanese and Korean). Supports nested block-level elements (such as P, DIV),

DVWA
Damn Vulnerable Web App (DVWA) is a PHP/MySQL web application that is very vulnerable. Its main goals are to be an aid for security professionals to test their skills and tools in a legal environment, to help web developers better understand the process of securing web applications, and to help teachers/students teach/learn in a classroom environment Web application security. The goal of DVWA is to practice some of the most common web vulnerabilities through a simple and straightforward interface, with varying degrees of difficulty. Please note that this software

SecLists
SecLists is the ultimate security tester's companion. It is a collection of various types of lists that are frequently used during security assessments, all in one place. SecLists helps make security testing more efficient and productive by conveniently providing all the lists a security tester might need. List types include usernames, passwords, URLs, fuzzing payloads, sensitive data patterns, web shells, and more. The tester can simply pull this repository onto a new test machine and he will have access to every type of list he needs.

Notepad++7.3.1
Easy-to-use and free code editor

MinGW - Minimalist GNU for Windows
This project is in the process of being migrated to osdn.net/projects/mingw, you can continue to follow us there. MinGW: A native Windows port of the GNU Compiler Collection (GCC), freely distributable import libraries and header files for building native Windows applications; includes extensions to the MSVC runtime to support C99 functionality. All MinGW software can run on 64-bit Windows platforms.