CAMEL-Bench: A Comprehensive Arabic LMM Benchmark
This addresses the problem of English-centric bias in LMM evaluation for over 400 million Arabic speakers, though it is incremental as it extends existing benchmarking approaches to a new language.
The authors tackled the lack of Arabic-language benchmarks for large multimodal models (LMMs) by developing CAMEL-Bench, a comprehensive evaluation benchmark covering eight diverse domains with 29,036 manually verified questions, and found that even top models like GPT-4o only achieved an overall score of 62%, indicating significant room for improvement.
Recent years have witnessed a significant interest in developing large multimodal models (LMMs) capable of performing various visual reasoning and understanding tasks. This has led to the introduction of multiple LMM benchmarks to evaluate LMMs on different tasks. However, most existing LMM evaluation benchmarks are predominantly English-centric. In this work, we develop a comprehensive LMM evaluation benchmark for the Arabic language to represent a large population of over 400 million speakers. The proposed benchmark, named CAMEL-Bench, comprises eight diverse domains and 38 sub-domains including, multi-image understanding, complex visual perception, handwritten document understanding, video understanding, medical imaging, plant diseases, and remote sensing-based land use understanding to evaluate broad scenario generalizability. Our CAMEL-Bench comprises around 29,036 questions that are filtered from a larger pool of samples, where the quality is manually verified by native speakers to ensure reliable model assessment. We conduct evaluations of both closed-source, including GPT-4 series, and open-source LMMs. Our analysis reveals the need for substantial improvement, especially among the best open-source models, with even the closed-source GPT-4o achieving an overall score of 62%. Our benchmark and evaluation scripts are open-sourced.