HMU-Bench：分层音乐理解基准
面向全能模型评测符号乐谱与音频音乐理解能力。
为什么重要
为音乐理解模型提供区分记谱与音频能力的评测依据。
阅读原文arxiv.org
AI 解读（根据原文生成）
论文提出 HMU-Bench，一个包含记谱与音频组件的中英双语音乐理解基准。ABC Q&A 含 1,029 个中英问题对，覆盖六个音乐领域，并标注为抽取、单步推理或多步推理；Audio Q&A 含 92 段录音上的 460 个问题对。作者用参考答案评测七个 omni models，发现强系统在记谱任务得分较高，但没有模型在 0-4 音频量表上平均超过 1.4。数据、标注、评测提示和打分响应将发布到 GitHub。
要点（AI 据原文整理，数字均已与原文核对）
- ABC Q&A contains 1,029 Chinese-English question pairs across six musical domains, labeled as extraction, single-step inference, or multi-step inference.
- Audio Q&A contains 460 question pairs on 92 recordings, spanning five scope-difficulty configurations.
- All music is Suno-generated, and questions and references are human-reviewed.
- Reference-based evaluation of seven omni models yields high notation scores for the strongest systems, but no model averages above 1.4 on the 0-4 audio scale.
- Notation and audio rankings show little correlation.
- A transposition case study also reveals octave errors missed by LLM grading.