MUUNRiver-Bench: Diagnosing Relation-Dependent Music Retrieval with Multimodal Instructions
Not on the map. Why?
Abstract (arXiv)
Music retrieval is relation-dependent: given a reference track, a listener may seek its style with a new theme, a cover, or a comparable voice, and these intents demand contradictory rankings. We present MUUNRiver-Bench, a diagnostic benchmark whose reference-audio queries use natural-language instructions to define relevance. A pipeline combining expert genre priors, LLM-generated prompts and lyrics, synthesis, and expert review yields 3,440 tracks spanning 13 genres and 116 sub-genres, and seven tasks: similar-music, style-preserving lyric-rewriting, lyric-preserving style-rewriting, cover, vocal-timbre, isolated-vocal, and segment retrieval. Across six models in eight configurations, task-wise rank reversals reveal complementary biases: acoustic encoders favour local identity, whereas text-aligned encoders favour semantic relations. Frozen encoders diagnose default similarity preferences; instruction-aware and audio-text fusion systems provide exploratory tests of textual conditioning, with neither simple fusion scheme consistently improving its backbone
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 16 of its 23 references that are in the map. First posted on arXiv 14 Sep 2026.
Builds on
16 of 23 references.
References · 16
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
- 2022Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentationcited by 719
- 2023MusicLM: Generating Music From Textcited by 448
- 2023MERT: Acoustic Music Understanding Model with Large-Scale Self-supervised Trainingcited by 181
- 2024GME: Improving Universal Multimodal Retrieval by Multimodal LLMscited by 170
- 2023UniIR: Training and Benchmarking Universal Multimodal Information Retrieverscited by 166
- 2022MuLan: A Joint Embedding of Music Audio and Natural Languagecited by 87
- 2023The Song Describer Dataset: a Corpus of Audio Captions for Music-and-Language Evaluationcited by 68
- 2025MuQ: Self-Supervised Music Representation Learning with Mel Residual Vector Quantizationcited by 68
- 2025CLaMP 3: Universal Music Information Retrieval Across Unaligned Modalities and Unseen Languagescited by 21
- 2024MidiCaps: A large-scale MIDI dataset with text captionscited by 19
- 2025CMI-Bench: A Comprehensive Benchmark for Evaluating Music Instruction Followingcited by 17
4 more
- 2024FakeMusicCaps: a Dataset for Detection and Attribution of Synthetic Music Generated via Text-to-Music Modelscited by 13
- 2026Muse: Towards Reproducible Long-Form Song Generation with Fine-Grained Style Controlcited by 8
- 2025WAVE: Learning Unified & Versatile Audio-Visual Embeddings with Multimodal LLMcited by 7
- 2026Efficient and High-Fidelity Omni Modality Retrievalcited by 3
Built on by
Not cited yet.