GAMEBoT: Transparent Assessment of LLM Reasoning in Games

TL;DR

GAMEBoT introduces a transparent, modular gaming benchmark for evaluating LLM reasoning, enabling detailed assessment of intermediate steps and reducing data contamination, thus providing a more rigorous measure of LLM capabilities.

Contribution

This paper presents GAMEBoT, a novel gaming benchmark that decomposes complex reasoning into subproblems, uses rule-based ground truth, and facilitates transparent, rigorous LLM evaluation.

Findings

Benchmark challenges LLM reasoning abilities.

LLMs struggle even with detailed CoT prompts.

GAMEBoT reduces data contamination risks.

Abstract

Large Language Models (LLMs) are increasingly deployed in real-world applications that demand complex reasoning. To track progress, robust benchmarks are required to evaluate their capabilities beyond superficial pattern recognition. However, current LLM reasoning benchmarks often face challenges such as insufficient interpretability, performance saturation or data contamination. To address these challenges, we introduce GAMEBoT, a gaming arena designed for rigorous and transparent assessment of LLM reasoning capabilities. GAMEBoT decomposes complex reasoning in games into predefined modular subproblems. This decomposition allows us to design a suite of Chain-of-Thought (CoT) prompts that leverage domain knowledge to guide LLMs in addressing these subproblems before action selection. Furthermore, we develop a suite of rule-based algorithms to generate ground truth for these subproblems, enabling rigorous validation of the LLMs' intermediate reasoning steps. This approach facilitates evaluation of both the quality of final actions and the accuracy of the underlying reasoning process. GAMEBoT also naturally alleviates the risk of data contamination through dynamic games and head-to-head LLM competitions. We benchmark 17 prominent LLMs across eight games, encompassing various strategic abilities and game characteristics. Our results suggest that GAMEBoT presents a significant challenge, even when LLMs are provided with detailed CoT prompts. Project page: https://visual-ai.github.io/gamebot