VoxSafeBench reveals that speech language models recognize social norms from text but fail to apply them when acoustic cues like speaker or scene determine the appropriate response.
MIT press, 2023
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
A statistical framework providing quantitative bounds on the effectiveness of multiple collectives steering classifier behavior via coordinated data modifications.
citing papers explorer
-
VoxSafeBench: Not Just What Is Said, but Who, How, and Where
VoxSafeBench reveals that speech language models recognize social norms from text but fail to apply them when acoustic cues like speaker or scene determine the appropriate response.
-
A Statistical Framework for Algorithmic Collective Action with Multiple Collectives
A statistical framework providing quantitative bounds on the effectiveness of multiple collectives steering classifier behavior via coordinated data modifications.