MM-Eval: A Multilingual Meta-Evaluation Benchmark for LLM-as-a-Judge and Reward Models
Son, Guijin, Yoon, Dongkeun, Suk, Juyoung, Aula-Blasco, Javier, Aslan, Mano, Kim, Vu Trong, Islam, Shayekh Bin, Prats-Cristià, Jaume, Tormo-Bañuelos, Lucía, Kim, Seungone
–arXiv.org Artificial Intelligence
Large language models (LLMs) are commonly used as evaluators in tasks (e.g., reward modeling, LLM-as-a-judge), where they act as proxies for human preferences or judgments. This leads to the need for meta-evaluation: evaluating the credibility of LLMs as evaluators. However, existing benchmarks primarily focus on English, offering limited insight into LLMs' effectiveness as evaluators in non-English contexts. To address this, we introduce MM-Eval, a multilingual meta-evaluation benchmark that covers 18 languages across six categories. MM-Eval evaluates various dimensions, including language-specific challenges like linguistics and language hallucinations. Evaluation results show that both proprietary and open-source language models have considerable room for improvement. Further analysis reveals a tendency for these models to assign middle-ground scores to low-resource languages. We publicly release our benchmark and code.
arXiv.org Artificial Intelligence
Oct-23-2024
- Country:
- Asia (1.00)
- Europe > Middle East
- Malta (0.14)
- North America > Mexico
- Mexico City (0.14)
- Genre:
- Research Report > New Finding (0.87)
- Industry:
- Education (1.00)
- Transportation (0.68)
- Technology: