Benchmark RadarAI BENCHMARK PROFILE
MRMAD
Evaluates LALMs on audio degradation perception through multi-turn dialogues over multiple audio inputs, covering speech, music, and sound.
- Released
- 2026-08-23
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Audio degradation perception is critical for real-world audio understanding, yet existing benchmarks overlook this low-level diagnostic capability.
Motivation
Large audio-language models (LALMs) have shown promising progress in understanding speech, music, and general sound events, yet their ability to reason about how audio signals are degraded remains underexplored.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.