Think Offline, Act in Real Time: Self-Improving Visual Game Agents through Code

Avatar
Poster
Voice is AI-generated
Connected to paperThis paper is a preprint and has not been certified by peer review

Think Offline, Act in Real Time: Self-Improving Visual Game Agents through Code

Authors

Zipeng Qiu, Shusheng Xu, Jiale Li, Yukun Zhou, Chenyue Li, Wenshuang Li, Tongkai Yang, Binhang Yuan

Abstract

Pre-trained models can read screens and reason about games, yet on an unfamiliar game, an agent still has to learn to play. We ask how an agent can self-improve on such a game from a few of its own attempts, in real-time visual games where the agent sees only the screen and the game keeps running while the agent thinks. We study two kinds of policy that can carry this improvement. In model as policy, a model selects actions during play; we adapt real-time chunking so that committed actions keep executing while the model reasons, and train a 2B controller for this interface. In code as policy, a model writes, tests, and revises a program between attempts, and the program acts during play. Across 170 GameWorld tasks, with at most eight one-minute executions per run, code as policy reaches 47.80% mean progress, against 30.90% for our trained controller, and leads in all five game categories. Improving the controller's weights on a single game, by contrast, yields only modest gains. These results favor letting a model think offline and storing what it learns in a program that acts in real time. Project repository: https://github.com/TabibitoQZP/SIGame

Follow Us on

0 comments

Add comment