Audio-language models retain 60-72% of benchmark scores without audio, and most audio-dependent items can be solved from short fragments rather than full clips.
Measuring audio’s impact on correctness: Audio-contribution-aware post-training of large audio language models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
All That Glitters Is Not Audio: Rethinking Text Priors and Audio Reliance in Audio-Language Evaluation
Audio-language models retain 60-72% of benchmark scores without audio, and most audio-dependent items can be solved from short fragments rather than full clips.