SDGBiasBench reveals intrinsic SDG biases in VLMs driven by priors rather than evidence, and CADE mitigates them with up to 25% accuracy gains and 12-point MAE reductions.
VisBias: Measuring Explicit and Implicit Social Biases in Vision Language Models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
The authors create a synthetic video auditing framework that detects statistically significant skin color biases in popular human action recognition models even when actions are identical.
citing papers explorer
-
SDGBiasBench: Benchmarking and Mitigating Vision--Language Models' Biases in Sustainable Development Goals
SDGBiasBench reveals intrinsic SDG biases in VLMs driven by priors rather than evidence, and CADE mitigates them with up to 25% accuracy gains and 12-point MAE reductions.
-
Identifying Ethical Biases in Action Recognition Models
The authors create a synthetic video auditing framework that detects statistically significant skin color biases in popular human action recognition models even when actions are identical.