AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Music Flamingo: Scaling Music Understanding in Audio Language Models

Sreyan Ghosh, Arushi Goel, Lasha Koroshinadze, Sang-gil Lee, Zhifeng Kong, Joao Felipe Santos, Ramani Duraiswami, Dinesh Manocha, Wei Ping, Mohammad Shoeybi, Bryan Catanzaro

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Speech and audio language models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add GRPO reinforcement learning with thinking traces after SFT (ablation)Music Flamingo · MMAU-Pro-Musicaccuracy: 63.9 -> 65.6Improved
add GRPO reinforcement learning with thinking traces after SFT (ablation)Music Flamingo · MuChoMusicaccuracy: 69.5 -> 74.58Improved
fine-tune Audio Flamingo 3 on MF-Skills and music dataAudio Flamingo 3 · MMAU Music full-testaccuracy: 73.95 -> 76.83Improved
fine-tune Audio Flamingo 3 on MF-Skills and music dataAudio Flamingo 3 · Music AVQAaccuracy: 76.7 -> 73.6Hurt
fine-tune Audio Flamingo 3 on MF-Skills and music dataAudio Flamingo 3 · SongCaps human ratingscore (1-10): 6.5 -> 8.3Improved
fine-tune Audio Flamingo 3 on MF-Skills and music dataAudio Flamingo 3 · NSynth Sourceaccuracy: 65.5 -> 75.89Improved
use a music-biased encoder (MERT) instead of captioning-trained Whisper encoder (ablation)linear probe · GS key classificationaccuracy: 40.56 -> 56.12Improved
use a music-biased encoder (MERT) instead of captioning-trained Whisper encoder (ablation)linear probe · GTZAN genre classificationaccuracy: 91.37 -> 78.96Hurt
add structured thinking reward and accuracy reward on top of format reward in GRPOMusic Flamingo GRPO · music caption and QA trainingreasoning qualityNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy