Human preferences for the same semantic content show near-chance agreement between text and audio, with audio raters using narrower decision thresholds, less length bias, and more user-oriented criteria.
Scaling laws for reward model overoptimization
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
RedDiffuser is a reinforced diffusion framework that generates adversarial visual contexts to audit and expose widespread multimodal safety failures in VLMs, increasing unsafe response rates by up to 10.69% on LLaVA with transfer to other models.
citing papers explorer
-
Same Words, Different Judgments: How Preferences Vary Across Modalities
Human preferences for the same semantic content show near-chance agreement between text and audio, with audio raters using narrower decision thresholds, less length bias, and more user-oriented criteria.
-
RedDiffuser: Auditing Multimodal Safety Failures in Vision-Language Models via Reinforced Diffusion
RedDiffuser is a reinforced diffusion framework that generates adversarial visual contexts to audit and expose widespread multimodal safety failures in VLMs, increasing unsafe response rates by up to 10.69% on LLaVA with transfer to other models.