Local Ai
Beyond Answer Changes: Matched Same-Class Controls for Local Audio Evidence
arXiv:2610.37605v1 Announce Type: cross Abstract: For audio-language models, an answer change after a local sound-event edit can reflect a response to the event's class, sensitivity to waveform rewrit
arXiv:2610.37605v1 Announce Type: cross Abstract: For audio-language models, an answer change after a local sound-event edit can reflect a response to the event's class, sensitivity to waveform rewriting, or both. We compare matched same-class and different-class replacements of a queried event. Both use the same interval, level estimate, donor fitting, and crossfade, while the same-class replacement preserves the correct answer. Across 600 synthetic soundscapes and three models from two architecture families, different-class replacements produce 49.7 to 60.9 percentage points more answer changes than same-class replacements. Same-class accuracy remains within 5.2 percentage points of original accuracy. Among trials with correct original and same-class answers, soundscape-weighted selective accuracy for the inserted class is 73.7 to 81.9%, averaging eligible trials within each soundscape and then weighting eligible soundscapes equally. This protocol tests how model answers respond to local changes in event class.
Related
- Event Fields: Learning Latent Event Structure for Waveform Foundation Models
- Represented but Ignored: A Causal Account of Prosodic Underuse in Audio-Language Models
- EASE-TTT: Evidence-Aligned Selective Test-Time Training for Long-Context Question Answering
Source: arXiv cs.AI | 2026-10-28