Bias

Your weight: normal

0.

Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling (arxiv.org)

0 points 1 sources 1 minutes ago cluster

Researchers propose a method to reduce bias in multimodal large language models (LLMs) by introducing perceptual perturbation and reward modeling. The approach aims to improve the fairness and accuracy of LLMs in judgment tasks.

bias llm models multimodal perceptual-perturbation
0.

Vision-Language Models Suppress Female Representations Under Ambiguous Input (arxiv.org)

0 points 1 sources 1 minutes ago cluster

Researchers found that vision-language models tend to suppress female representations when given ambiguous input, according to a study published on arXiv. The study analyzed the performance of these models on tasks involving gender classification.

bias computer-vision models
0.

Algorithmic Monocultures in Hiring (arxiv.org)

0 points 1 sources 1 minutes ago cluster

Researchers from Stanford University published a paper on arXiv AI exploring how AI-driven hiring systems perpetuate existing biases. The study examines the impact of algorithmic monocultures in hiring processes.

ai bias hiring