Topic Bias

Topic bias in artificial intelligence (AI) refers to the systematic errors in AI models stemming from skewed training data or algorithmic design, leading to unfair or inaccurate outputs for certain groups or topics. Current research focuses on identifying and mitigating these biases across various AI applications, including language models (LLMs), image recognition systems, and even models used in healthcare and social sciences, employing techniques like adversarial training, fairness-aware algorithms, and bias detection frameworks. Understanding and addressing topic bias is crucial for ensuring fairness, reliability, and ethical deployment of AI systems across diverse contexts, impacting both the development of more robust AI and the equitable application of AI-driven technologies in society.

Papers

September 9, 2023

Correcting sampling biases via importance reweighting for spatial modeling
Boris Prokhorov, Diana Koldasbayeva, Alexey Zaytsev
Machine Learning Model Estimation Task Language Correction Topic Bias Distributional Assumption Kernel Density Distribution Shift Problem Spatial Modelling

August 24, 2023

REB: Reducing Biases in Representation for Industrial Anomaly Detection
Shuai Lyu, Dongmei Mo, Waikeung Wong
Anomaly Detection Individual Representation Topic Bias Industrial Anomaly Detection Bias Reduction Domain Bias

August 21, 2023

FairMonitor: A Four-Stage Automatic Framework for Detecting Stereotypes and Biases in Large Language Models
Yanhong Bai, Jiabao Zhao, Jinxin Shi, Tingjiang Wei, Xingjiao Wu, Liang He
Implicit Bias Stereotype Content Topic Bias Multi Dimensional Evaluation Stereotype Detection Multi Stage Framework

August 18, 2023

From Hope to Safety: Unlearning Biases of Deep Models via Gradient Penalization in Latent Space
Maximilian Dreyer, Frederik Pahde, Christopher J. Anders, Wojciech Samek, Sebastian Lapuschkin
Latent Space Deep Model Topic Bias Gradient Regularization Concept Activation Vector Model Sensitivity Hope Speech

June 27, 2023

Gender Bias in BERT -- Measuring and Analysing Biases through Sentiment Rating in a Realistic Downstream Classification Task
Sophie Jentzsch, Cigdem Turan
BERT Model Gender Bias Ticket BERT Topic Bias Sentiment Score Downstream Classification

June 7, 2023

An Overview of Challenges in Egocentric Text-Video Retrieval
Burak Satar, Hongyuan Zhu, Hanwang Zhang, Joo Hwee Lim
Technical Challenge Egocentric Video Topic Bias Video Text Retrieval Multiple Source Length Bias

May 26, 2023

Finspector: A Human-Centered Visual Inspection Tool for Exploring and Comparing Biases among Foundation Models
Bum Chul Kwon, Nandana Mihindukulasooriya
Language Model Foundation Model Topic Bias Visual Inspection Transformer Based Pre Trained Language Potential Bias

May 24, 2023

In-Context Impersonation Reveals Large Language Models' Strengths and Biases
Leonard Salewski, Stephan Alaniz, Isabel Rio-Torto, Eric Schulz, Zeynep Akata
Topic Bias Estimated Team Strength Diverse Persona Context Impersonation

May 18, 2023

Comparing Biases and the Impact of Multilingual Training across Multiple Languages
Sharon Levy, Neha Anna John, Ling Liu, Yogarshi Vyas, Jie Ma, Yoshinari Fujinuma, Miguel Ballesteros, Vittorio Castelli, Dan Roth
Global Impact Absolute Stance Bias Social Bias Topic Bias Bias Evaluation Multiple Language Multilingual Training

May 17, 2023

May 10, 2023

A Classification of Feedback Loops and Their Relation to Biases in Automated Decision-Making Systems
Nicolò Pagan, Joachim Baumann, Ezzat Elokda, Giulia De Pasquale, Saverio Bolognani, Anikó Hannák
Classification Code Topic Bias Relational Information Feedback Loop Automated Decision Making Feedback Effect

May 3, 2023

Maximizing Submodular Functions for Recommendation in the Presence of Biases
Anay Mehrotra, Nisheeth K. Vishnoi
Effective Recommendation Speech Presence Submodular Maximization Topic Bias Near Optimal Balancing Efficiency

April 28, 2023

Information Redundancy and Biases in Public Document Information Extraction Benchmarks
Seif Laatiri, Pirashanth Ratnamogan, Joel Tang, Laurent Lam, William Vanhuffel, Fabien Caspani
Information Redundancy Topic Bias Visually Rich Document Human Annotated Open Information Extraction Key Information Extraction Document Understanding Task

April 3, 2023

Design Choices for Crowdsourcing Implicit Discourse Relations: Revealing the Biases Introduced by Task Design
Valentina Pyatkin, Frances Yung, Merel C. J. Scholman, Reut Tsarfaty, Ido Dagan, Vera Demberg
Topic Bias Annotation Task Design Choice Implicit Discourse Relation Linguistic Annotation Natural Language Annotation

March 2, 2023

Penalising the biases in norm regularisation enforces sparsity
Etienne Boursier, Nicolas Flammarion
Regularization Model Sparsity Increase Topic Bias Layer ReLU Network Sparse Solution Sparsity Inducing Minimum Norm

February 10, 2023

FairPy: A Toolkit for Evaluation of Social Biases and their Mitigation in Large Language Models
Hrishikesh Viswanath, Tianyi Zhang
Language Model Global Evaluation Social Bias Topic Bias Easy to Use Toolkit Quantitative Evaluation

January 18, 2023

Biases in Scholarly Recommender Systems: Impact, Prevalence, and Mitigation
Michael Färber, Melissa Coutinho, Shuzhou Yuan
Global Impact Recommender System Faulty Negative Mitigation Topic Bias Prevalence Threshold Publication Bias

January 14, 2023

The Role of Heuristics and Biases During Complex Choices with an AI Teammate
Nikolos Gurney, John H. Miller, David V. Pynadath
Artificial Intelligence Integral Role Decision Making Topic Bias Simple Heuristic AI Assistance Human AI Decision Making AI Teammate Algorithmic Decision Support

January 13, 2023

Biases in Inverse Ising Estimates of Near-Critical Behaviour
Maximilian Benedikt Kloucek, Thomas Machon, Shogo Kajimura, C. Patrick Royall, Naoki Masuda, Francesco Turci
Functional Magnetic Resonance Imaging Topic Bias Ising Model Criticality Ordered Spin Sequence Pseudo Likelihood Criticality Analysis

Topic Bias

Papers

Correcting sampling biases via importance reweighting for spatial modeling

REB: Reducing Biases in Representation for Industrial Anomaly Detection

FairMonitor: A Four-Stage Automatic Framework for Detecting Stereotypes and Biases in Large Language Models

From Hope to Safety: Unlearning Biases of Deep Models via Gradient Penalization in Latent Space

Gender Bias in BERT -- Measuring and Analysing Biases through Sentiment Rating in a Realistic Downstream Classification Task

An Overview of Challenges in Egocentric Text-Video Retrieval

Finspector: A Human-Centered Visual Inspection Tool for Exploring and Comparing Biases among Foundation Models

In-Context Impersonation Reveals Large Language Models' Strengths and Biases

Comparing Biases and the Impact of Multilingual Training across Multiple Languages

Echoes of Biases: How Stigmatizing Language Affects AI Performance

"I'm fully who I am": Towards Centering Transgender and Non-Binary Voices to Measure Biases in Open Language Generation

A Classification of Feedback Loops and Their Relation to Biases in Automated Decision-Making Systems

Maximizing Submodular Functions for Recommendation in the Presence of Biases

Information Redundancy and Biases in Public Document Information Extraction Benchmarks

Design Choices for Crowdsourcing Implicit Discourse Relations: Revealing the Biases Introduced by Task Design

Penalising the biases in norm regularisation enforces sparsity

FairPy: A Toolkit for Evaluation of Social Biases and their Mitigation in Large Language Models

Biases in Scholarly Recommender Systems: Impact, Prevalence, and Mitigation

The Role of Heuristics and Biases During Complex Choices with an AI Teammate

Biases in Inverse Ising Estimates of Near-Critical Behaviour