Beyond Text Following: Repairable Arbitration Reversals in Audio-Language Models

Published
Source
arXiv
Paper number
307
Field
LLMs / NLP
arXiv ID
2606.05161

Key points

  • The paper examines this question using an audio-fixed counterfactual method that removes only conflicting text and measures the resulting change in model preferences.
  • Across five ALMs and four conflict tasks, sign flips appeared in 64.1 percent of conflict samples. In other words, the same audio-only branch preferred the audio-supported answer, while the combined branch preferred the text-supported answer.
  • Under a strict 5 percentage point fidelity-drop budget, GACL improves nAUC by 17.8 points over the best contrastive-learning baseline and transfers to vision-text mediation without retraining, with up to a 40.5 percentage point gain.

Paper links

External research summaries. These are not HDATF publications or measured product results.

Read original (opens in a new tab)