The Flow has not summarised this story yet — read it at arXiv Computation and Language.
Project Qualia investigates whether experiential similarity between songs can be extracted from listening behavior. Using 1.29 billion scrobbles from 9,396 users, the authors trained a Word2Vec model (Song2Vec) on session data, then applied an artist‑residual procedure to isolate artist‑independent signals. The residual embeddings still contained strong cross‑artist similarity, forming coherent genre and era clusters, demonstrating that experiential structure exists beyond artist identity.
The paper investigates whether expert critic‑sourced relationships between musical artists—called critical adjacency—can be recovered from the artists’ acoustic content. By modeling artists as distributions over 80 low‑level acoustic descriptors and measuring pairwise proximity with Wasserstein distances, the authors achieve an out‑of‑sample AUC of 0.767 for predicting critic‑linked edges, with higher recoverability for edges supported by multiple critics. The results suggest that critical discourse contains a reproducible sonic core, while sociological factors also influence the perceived adjacency between artists.
arXiv:2601.11262v2 Announce Type: replace-cross Abstract: Music Cover Retrieval, also known as Version Identification, aims to recognize distinct renditions of the same underlying musical work, a tas...
The paper introduces the MATCHA dataset, comprising 1,105 perceptual assessments from 83 experts on attribute-based music matches across five musical attributes—melody, harmony, rhythm, voice, and timbre. A triplet-based forced-choice experiment with 300 cases, including plagiarism, cover songs, and AI-generated music, was used to gather these judgments. Results show measurable agreement among participants and partial alignment with computational similarity measures, highlighting the need for perceptually grounded evaluation in generative AI for music.
The paper evaluates Graph Neural Networks (GNNs) for predicting artist success within collaboration networks, extending prior work on Italian and Danish music scenes by adding a Polish dataset and merging the three into a tri‑national network. Statistical analysis shows the Polish and combined networks share similar clustering properties, while predictive experiments reveal that GNNs match or slightly outperform a Multilayer Perceptron (MLP) in some cases but the MLP generally yields higher success metrics. The findings suggest that internal node attributes such as genre and label affiliation may be more predictive than network topology, and that GNNs may better capture cross‑border relational structures in the merged network.
MUUNRiver-Bench is a diagnostic benchmark for music retrieval that uses natural‑language instructions to define relevance for reference‑audio queries. It contains 3,440 tracks across 13 genres and 116 sub‑genres and covers seven tasks such as similar‑music, style‑preserving lyric‑rewriting, cover, and segment retrieval. Experiments with six models in eight configurations show that acoustic encoders favor local identity while text‑aligned encoders favor semantic relations, and that instruction‑aware and audio‑text fusion systems do not consistently outperform their backbones.