LLM agent Porimon improves Pokémon battles with knowledge augmentation
Porimon: An LLM-Based Pokémon Battle Agent Enhanced by Long/Short-Term Knowledge Augmented Generation
Artificial Intelligence
Summary
Playing Pokémon battles well means planning moves while thinking about your opponent's strategy. The authors developed Porimon, an AI agent powered by large language models (LLMs), which uses a new method called Long/Short-Term Knowledge Augmented Generation (LSTKAG) to remember and learn from past battles and game states. Porimon also uses a tool to calculate exact damage and game details to make better decisions. Tests show Porimon beats previous AI players and simple rule-based opponents, especially when it uses updated game information. However, the benefit of recalling long-term past experiences is less clear.
What this means in practice
- •For game ai developers: Build AI agents for turn-based games that improve opponents’ strategy awareness through knowledge of past moves and detailed state information.
- •For interactive entertainment software teams: Enhance NPC behavior in player-versus-player games by integrating memory-augmented language models and precise game-state APIs for strategic planning.
Authors
Dongyin Zhuo, Fengjunjie Pan, Nenad Petrovic, Alois Knoll
Abstract
In this paper, we use Pokémon Battles as a case study to investigate how to improve the performance of LLM-based agents in tasks that require opponent-aware planning without additional fine-tuning. We propose Long/Short-Term Knowledge Augmented Generation (LSTKAG), a mechanism that enables LLM-based agents to leverage past states of the current task and retrieve experience summaries from similar previous task instances based on the current state. Based on LSTKAG, we design Porimon, an LLM-based agent structure for Pokémon Battles. For optimization, we introduce an external API for precise damage calculation and more detailed information about the game. We conduct tournament-like evaluation experiments comprising 15,000 battles for hyperparameter optimization, ablation studies, and performance evaluation. The results indicate that Porimon-based players with hyperparameter optimization significantly outperform players based on PokéLLMon, an LLM-based agent structure proposed in previous research, and the rule-based heuristic player. Furthermore, our ablation study shows that Porimon variants outperform the one without extension in game information retrieval, which shows the contribution of that extension. However, the current experiment results are inconclusive regarding the contribution of Long-Term KAG. These results suggest that introducing external resources, information from previous states of the current task, and experience summaries from similar previous task instances could elevate the performance of LLM-based agents designed for tasks requiring opponent-aware planning.