Bias Score

Bias scores are quantitative metrics designed to assess and quantify biases present in various machine learning models and datasets, aiming to promote fairness and mitigate discriminatory outcomes. Current research focuses on developing novel bias score methodologies, particularly for text data (e.g., analyzing language models and documents) and image data (e.g., in medical image analysis), often employing techniques like topic modeling, word embeddings, and deep learning architectures such as transformers. These advancements are crucial for improving the reliability and ethical implications of AI systems across diverse applications, from news classification and healthcare to legal and scientific domains, by enabling the identification and mitigation of biases that might otherwise lead to unfair or inaccurate predictions.

Papers

November 13, 2024

Impact of Iris Pigmentation on Performance Bias in Visible Iris Verification Systems: A Comparative Study
Geetanjali Sharma, Abhishek Tandon, Gaurav Jaswal, Aditya Nigam, Raghavendra Ramachandra
Comparative Study Iris Recognition Bias Score Iris Pattern

October 29, 2024

Attention Speaks Volumes: Localizing and Mitigating Bias in Language Models
Rishabh Adiga, Besmira Nushi, Varun Chandrasekaran
Full Model Human Attention Human Language Absolute Stance Bias Bias Mitigation Attention Layer Mitigating Bias Automatic Volume Bias Score Stochastic Layer

October 17, 2024

LLMs are Biased Teachers: Evaluating LLM Bias in Personalized Education
Iain Weissburg, Sathvika Anand, Sharon Levy, Haewon Jeong
Large Language Model Inherent Bias LLM Bias Bias Score Source Bias Personalized Education

September 24, 2024

Beats of Bias: Analyzing Lyrics with Topic Modeling and Gender Bias Measurements
Danqing Chen, Adithi Satish, Rasul Khanbayov, Carolin M. Schuster, Georg Groh
Absolute Stance Bias Gender Bias Topic Modeling Rhythm Game Song Lyric Bias Measurement Bias Score

July 19, 2024

LLMs left, right, and center: Assessing GPT's capabilities to label political bias from web domains
Raphael Hernandes, Giulio Corsi
Medical LLM GPT Neo Capability Evolution Domain Name Political Bias Medium Bias Dominance Hierarchy Bias Score

July 11, 2024

BiasPruner: Debiased Continual Learning for Medical Image Classification
Nourhan Bayasi, Jamil Fayyad, Alceu Bissoto, Ghassan Hamarneh, Rafeef Garbi
Continual LEArning Catastrophic Forgetting Medical Image Classification Sequential Learning Bias Score Bias Challenge

May 23, 2024

Subtle Biases Need Subtler Measures: Dual Metrics for Evaluating Representative and Affinity Bias in Large Language Models
Abhishek Kumar, Sarfaroz Yunusov, Ali Emami
Subtle Bias Bias Score Biased Association Machine Bias Affinity Score

May 8, 2024

Towards Less Biased Data-driven Scoring with Deep Learning-Based End-to-end Database Search in Tandem Mass Spectrometry
Yonghan Yu, Ming Li
Mass Spectrometry Bias Score Neural Search Spectrometry Based Proteomics Peptide Level Identification Precision

January 27, 2024

Semantic Properties of cosine based bias scores for word embeddings
Sarah Schröder, Alexander Schulz, Fabian Hinder, Barbara Hammer
Language Model Absolute Stance Bias Word Embeddings Social Bias Bias Detection Bias Score Semantic Property

October 25, 2023

Multilingual Coarse Political Stance Classification of Media. The Editorial Line of a ChatGPT and Bard Newspaper
Cristina España-Bonet
ChatGPT Generated Conversation Political Bias Medium Bias Ethnic Medium Bias Score Google Bard Political Stance

October 9, 2023

Post-hoc Bias Scoring Is Optimal For Fair Classification
Wenlong Chen, Yegor Klochkov, Yang Liu
Fairness Constraint Fair Classification Bias Score Group Fairness Constraint

September 25, 2023

(Predictable) Performance Bias in Unsupervised Anomaly Detection
Felix Meissen, Svenja Breuer, Moritz Knolle, Alena Buyx, Ruth Müller, Georgios Kaissis, Benedikt Wiestler, Daniel Rückert
Unsupervised Anomaly Detection Medical Image Data Predictability Problem Balanced Dataset Chest X Ray Datasets Bias Score

July 16, 2023

It's All Relative: Interpretable Models for Scoring Bias in Documents
Aswin Suresh, Chi-Hsuan Wu, Matthias Grossglauser
Inherent Interpretability Absolute Stance Bias Interpretable Model Document Relevance Bias Score

March 27, 2023

Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generation
Susung Hong, Donghoon Ahn, Seungryong Kim
Complex Prompt Text to 3D Generation Score Distillation Score Distillation Sampling Bias Score 2 Dimensional Diffusion

February 22, 2023

Delving into Identify-Emphasize Paradigm for Combating Unknown Bias
Bowen Zhao, Chen Chen, Qian-Wei Wang, Anfeng He, Shu-Tao Xia
Task Specific Model Dataset Bias Inherent Bias Bias Score Bias Aligned Sample

June 19, 2022

Predicting Human Performance in Vertical Hierarchical Menu Selection in Immersive AR Using Hand-gesture and Head-gaze
Majid Pourmemar, Yashas Joshi, Charalambos Poullis
Augmented Reality Hand Gesture User Study Human Performance Bias Score Visual Hierarchy Augmented Reality Application

March 28, 2022

The SAME score: Improved cosine based bias score for word embeddings
Sarah Schröder, Alexander Schulz, Barbara Hammer
Word Embeddings Sentence Embeddings Cosine Similarity Bias Score

November 25, 2021

Combating Unknown Bias with Effective Bias-Conflicting Scoring and Gradient Alignment
Bowen Zhao, Chen Chen, Qian-Wei Wang, Anfeng He, Shu-Tao Xia
Task Specific Model Dataset Bias Inherent Bias Gradient Alignment Bias Score Bias Aligned Sample

November 18, 2021

Assessing Social Determinants-Related Performance Bias of Machine Learning Models: A case of Hyperchloremia Prediction in ICU Population
Songzi Liu, Yuan Luo
Machine Learning Machine Learning Model Simple Classifier Health Disparity Social Determinant Bias Score Intensive Care Unit Mortality

November 15, 2021

Assessing gender bias in medical and scientific masked language models with StereoSet
Robert Robinson
Language Model Gender Bias Mono to Stereo Bias Score

Bias Score

Papers

Impact of Iris Pigmentation on Performance Bias in Visible Iris Verification Systems: A Comparative Study

Attention Speaks Volumes: Localizing and Mitigating Bias in Language Models

LLMs are Biased Teachers: Evaluating LLM Bias in Personalized Education

Beats of Bias: Analyzing Lyrics with Topic Modeling and Gender Bias Measurements

LLMs left, right, and center: Assessing GPT's capabilities to label political bias from web domains

BiasPruner: Debiased Continual Learning for Medical Image Classification

Subtle Biases Need Subtler Measures: Dual Metrics for Evaluating Representative and Affinity Bias in Large Language Models

Towards Less Biased Data-driven Scoring with Deep Learning-Based End-to-end Database Search in Tandem Mass Spectrometry

Semantic Properties of cosine based bias scores for word embeddings

Multilingual Coarse Political Stance Classification of Media. The Editorial Line of a ChatGPT and Bard Newspaper

Post-hoc Bias Scoring Is Optimal For Fair Classification

(Predictable) Performance Bias in Unsupervised Anomaly Detection

It's All Relative: Interpretable Models for Scoring Bias in Documents

Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generation

Delving into Identify-Emphasize Paradigm for Combating Unknown Bias

Predicting Human Performance in Vertical Hierarchical Menu Selection in Immersive AR Using Hand-gesture and Head-gaze

The SAME score: Improved cosine based bias score for word embeddings

Combating Unknown Bias with Effective Bias-Conflicting Scoring and Gradient Alignment

Assessing Social Determinants-Related Performance Bias of Machine Learning Models: A case of Hyperchloremia Prediction in ICU Population

Assessing gender bias in medical and scientific masked language models with StereoSet