MusicTwins：训练音频语言模型像特定用户那样听音乐
MusicTwins: Training Audio-Language Models to Listen like Their Persona
AI 导读
研究者提出 MusicTwins 框架，结合人格摘要、个人记忆和后训练，让音频语言模型生成特定用户的个性化回应。团队基于人类文字构建人格表征，通过监督微调和强化学习后训练 Audio Flamingo Next。在 10 个保留人格上的评估显示，模型在人格忠实度（包括性格、写作风格和对音乐的回应）方面显著提升。
正文 · 原文
Abstract:People differ in how they perceive music, what associations they make with it, what they like about it, and how they respond to it. Computational models intended to simulate people in environments that include music should account for these differences. Although LLMs have been shown to reproduce aspects of human behavior, whether audio-language models (ALMs) can reproduce a specific person's responses to textual and musical stimuli remains underexplored. We introduce MusicTwins, a framework that combines persona summaries, personal memory, and post-training to generate person-specific responses. We construct persona representations from human writings, capturing background, views, personality, writing style, and a musical profile, and post-train Audio Flamingo Next through supervised fine-tuning and reinforcement learning. Evaluation on ten held-out personas shows notable improvements for Audio Flamingo Next in persona fidelity, including personality, writing style, and responses to music, supporting MusicTwins as a foundation for training audio-language models to listen like their persona.
| Subjects: | Sound (cs.SD); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.32495 [cs.SD] |
| (or arXiv:2610.32495v1 [cs.SD] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32495 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Milan Liessens Dujardin [view email]
[v1]
Fri, 23 Oct 2026 11:11:45 UTC (2,513 KB)
来源：arXiv cs.AI · arxiv.org