CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents
This addresses the problem of limited and single-environment benchmarks for researchers and developers working on autonomous agents using Multimodal Language Models, though it is incremental as it builds on existing agent evaluation methods.
The authors tackled the lack of cross-environment benchmarks for Multimodal Language Model agents by introducing CRAB, a framework with a graph-based evaluation method, and achieved a best completion ratio of 38.01% with GPT-4o on 120 tasks across desktop and mobile environments.
The development of autonomous agents increasingly relies on Multimodal Language Models (MLMs) to perform tasks described in natural language with GUI environments, such as websites, desktop computers, or mobile phones. Existing benchmarks for MLM agents in interactive environments are limited by their focus on a single environment, lack of detailed and generalized evaluation methods, and the complexities of constructing tasks and evaluators. To overcome these limitations, we introduce Crab, the first agent benchmark framework designed to support cross-environment tasks, incorporating a graph-based fine-grained evaluation method and an efficient mechanism for task and evaluator construction. Our framework supports multiple devices and can be easily extended to any environment with a Python interface. Leveraging Crab, we developed a cross-platform Crab Benchmark-v0 comprising 120 tasks in computer desktop and mobile phone environments. We evaluated four advanced MLMs using different single and multi-agent system configurations on this benchmark. The experimental results demonstrate that the single agent with GPT-4o achieves the best completion ratio of 38.01%. All framework code, agent code, and task datasets are publicly available at https://github.com/camel-ai/crab.