CMI-Bench converts standard MIR annotations into instruction-following tasks and shows current audio-text LLMs underperform supervised MIR systems across nearly all 14 tasks.
Our results highlight a significant performance gap between LLMs and supervised MIR systems, with best models like Qwen2-Audio and GAMA also struggling with general- ization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CMI-Bench: A Comprehensive Benchmark for Evaluating Music Instruction Following
CMI-Bench converts standard MIR annotations into instruction-following tasks and shows current audio-text LLMs underperform supervised MIR systems across nearly all 14 tasks.