Generative Agents for Multi-Agent Autoformalization of Interaction Scenarios
This reduces the manual effort and domain expertise needed for developing multi-agent simulations, though it is incremental as it builds on existing LLM and validation techniques.
The paper tackles automating the formalization of multi-agent interaction scenarios from natural language descriptions into executable logic programs, achieving up to 100% syntactic and 77% semantic correctness in experiments with 110 descriptions across five game-theoretic scenarios.
Multi-agent simulations are versatile tools for exploring interactions among natural and artificial agents, but their development typically demands domain expertise and manual effort. This work introduces the Generative Agents for Multi-Agent Autoformalization (GAMA) framework, which automates the formalization of interaction scenarios in simulations using agents augmented with large language models (LLMs). To demonstrate the application of GAMA, we use natural language descriptions of game-theoretic scenarios representing social interactions, and we autoformalize them into executable logic programs defining game rules, with syntactic correctness enforced through a solver-based validation. To ensure runtime validity, an iterative, tournament-based procedure tests the generated rules and strategies, followed by exact semantic validation when ground truth outcomes are available. In experiments with 110 natural language descriptions across five 2x2 simultaneous-move games, GAMA achieves 100% syntactic and 76.5% semantic correctness with Claude 3.5 Sonnet, and 99.82% syntactic and 77% semantic correctness with GPT-4o. The framework also shows high semantic accuracy in autoformalizing agents' strategies.