A contrastive-style adapter trained on LLM-generated positive and negative audio descriptions improves audio hallucination accuracy to 77.5 percent and audio question answering to 84.3 percent, without changing the frozen language model.
It aims to generate descriptions of both the sound events that are present and those that are ab- sent in the audio
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
extension 1
citation-polarity summary
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1roles
extension 1polarities
extend 1representative citing papers
citing papers explorer
-
Teaching Audio-Aware Large Language Models What Does Not Hear: Mitigating Hallucinations through Synthesized Negative Samples
A contrastive-style adapter trained on LLM-generated positive and negative audio descriptions improves audio hallucination accuracy to 77.5 percent and audio question answering to 84.3 percent, without changing the frozen language model.