P-MMEval: A Parallel Multilingual Multitask Benchmark for Consistent Evaluation of LLMs
This provides a standardized evaluation framework for researchers and developers working on multilingual LLMs, though it is incremental as it builds on existing benchmarking approaches.
The authors tackled the lack of comprehensive multilingual multitask benchmarks for LLMs by introducing P-MMEval, a large-scale benchmark covering fundamental and specialized datasets with consistent language coverage and parallel samples, and conducted extensive experiments to compare model performances and explore factors like tasks, model sizes, languages, and knowledge transfer.
Recent advancements in large language models (LLMs) showcase varied multilingual capabilities across tasks like translation, code generation, and reasoning. Previous assessments often limited their scope to fundamental natural language processing (NLP) or isolated capability-specific tasks. To alleviate this drawback, we aim to present a comprehensive multilingual multitask benchmark. First, we introduce P-MMEval, a large-scale benchmark covering effective fundamental and capability-specialized datasets. Furthermore, P-MMEval delivers consistent language coverage across various datasets and provides parallel samples. Finally, we conduct extensive experiments on representative multilingual model series to compare performances across models and tasks, explore the relationship between multilingual performances and factors such as tasks, model sizes, languages, and prompts, and examine the effectiveness of knowledge transfer from English to other languages. The resulting insights are intended to offer valuable guidance for future research. The dataset is available at https://huggingface.co/datasets/Qwen/P-MMEval.