AlignBench: Benchmarking Chinese Alignment of Large Language Models

Alignment has become a critical step for instruction-tuned Large Language Models (LLMs) to become helpful assistants. However, the effective evaluation of alignment for emerging Chinese LLMs is still significantly lacking, calling for real-scenario grounded, open-ended, challenging and automatic evaluations tailored for alignment. To fill in this gap, we introduce ALIGNBENCH, a comprehensive multidimensional benchmark for evaluating LLMs' alignment in Chinese. We tailor a humanin-the-loop data curation pipeline, containing 8 main categories, 683 real-scenario rooted queries and corresponding human verified references. To ensure the correctness of references, each knowledge-intensive query is accompanied with evidences collected from reliable web sources (including URLs and quotations) by our annotators. For automatic evaluation, our benchmark employs a rule-calibrated multi-dimensional LLM-as-Judge (Zheng et al., 2023) approach with Chain-of-Thought to generate explanations and final ratings, ensuring high reliability and interpretability. All evaluation codes, data, and LLM generations are available at https://github.com/THUDM/ AlignBench .