Stereotype Mitigation

Follow this topic Ask about this topic

Momentum

9 papers in the last four weeks, up 200% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 42

All topics

Acmite: Mitigating Gender Bias in LLMs through Concept-Guided Mutual Information
Oct 1, 2026Tian Lan, Xiaoqing Cheng, Han Zhang +1Gender Bias Stereotype Mitigation
Kyoto University · Zhejiang University · Shanghai Jiao Tong University +1
When the Judge Acts: Auditing VLM-Guided Image Selection on Culturally Situated Prompts
Oct 1, 2026Huichan SeoLarge Language Model Judges Judges
Independent Researcher
Gender bias across LLMs is common and highly heterogeneous
Sep 29, 2026Edoardo Bolzoni, Valerio CapraroGender Bias Stereotype Mitigation
University of Milan-Bicocca, Milan, Italy
IMPLICIT-Bench: Measuring Implicit Bias in Text-to-Image Models under Neutral Prompts
Sep 21, 2026Yue Dai, Ziyang Liu, Marc Cheong +1Modern Text-To-Image Models Text-To-Image
The University of Melbourne
Deconstructing Stereotypes: Scope-Conditioned Generation for Effective Multilingual Counterspeech
Sep 15, 2026Greta Damo, Elias Urios Alacreu, Elena Cabrio +2Stereotype Mitigation Hate Speech
1Universit´e Cˆote d’Azur, CNRS, INRIA, I3S, France · 2PRHLT Research Center, Universitat Polit`ecnica de Val`encia, Valencia, Spain · 3ValgrAI Valencian Graduate School and Research Network of Artificial Intelligence, Spain
Prompt Revision as a Source of Cultural Bias in Text-to-Image Systems
Sep 12, 2026Aleksandra Urman, Elsa Lichtenegger, Salima Jaoua +7Stereotype Mitigation Text Generation
University of Zurich · University of Helsinki · ETH Zurich +1
When Noise Fabricates Bias: The Fragility of LLM-as-a-Judge Bias Measurement under Noisy Text
Sep 10, 2026DongHyun Ryu, Jaehyeok Lee, YeongJun Hwang +1Large Language Model Bias Llm-As-A-Judge
Sungkyunkwan University, Suwon, South Korea
Voice or Stereotype? Disentangling Acoustic and Content-Based Gender in Speech-to-Speech Models
Sep 10, 2026Xiaoqun Liu, Tanu Mitra, Harshit Rajgarhia +1Gender Seed-Tts-Eval Benchmark
Centific Research · University of Washington
Tracing Stereotypes from Representation to Output in Multilingual LLMs
Sep 8, 2026Ariun-Erdene Tumurchuluun, Yusser Al Ghussin, Pinzhen Chen +2Stereotype Mitigation Cross-Lingual Transfer
Saarland University, Saarland Informatics Campus · German Research Center for Artificial Intelligence (DFKI) · Queen’s University Belfast +1
Right Frame, Wrong Rule: Cultural Cues Expose the Financial Knowledge Gap They Were Meant to Close
Sep 1, 2026Rania Elbadry, Ahmed Heakl, Saeed Almheiri +12Social Norms Interpretive Layer
♠MBZUAI · ♢The University of Tokyo · §The Fin AI +4
Evaluating and Mitigating Anti-LGBTQ Biases in German and Multilingual Language Models
Aug 31, 2026Melina Morch, Daniel BraunLarge Language Model Bias Multilingual Benchmark
Marburg University · Computer Science · Cultural Data Studies +2
ContextBias: Controlled Evaluation of Bias Persistence Under Context Shift in Text-to-Image Models
Aug 30, 2026Shaghayegh Kolli, Sina Emami, Moreno D'Incà +4Modern Text-To-Image Models Stereotype Mitigation
1Technical University of Munich · 4Munich Center for Machine Learning (MCML) · 5Munich Data Science Institute (MDSI) +2
Weather- and Location-Aware Agentic Dining Recommendation: Leveraging LLM World Knowledge for Region-Sensitive Contextual Reasoning
Aug 5, 2026Kadharmoideen FadurudeenContextual Reasoning Recipe
Independent Researcher
STEREODISCO: Discovering Stereotypicality in LLMs
Jul 30, 2026Farane Jalali Farahani, Corina Dima, Mojtaba Nayyeri +2Stereotype Mitigation Large Language Model Annotations
Institute for Artificial Intelligence, Stuttgart, Germany · Institute for Social Science, Stuttgart, Germany · University of Southampton, Southampton, United Kingdom
Evaluating Regional Bias in LLMs From Abstract Stereotype to Concrete Social Decision-Making
Jul 29, 2026Jiayuan Di, Haoyi Yang, Yufei Luo +2Stereotype Mitigation Geometric Bias
East China University of Science and Technology · Mohamed bin Zayed University of Artificial Intelligence · Beijing University of Posts and Telecommunications +2
Probing Latent Colombian Identity Inferences in Qwen2.5-7B with Natural Language Autoencoders
Jul 23, 2026Pablo Santiago Potes Velasco, María del Mar García Matabanchoy, Óscar Julián Pérez Ladino +3Stereotype Mitigation Interpretability
Universidad Autónoma de Occidente Cali, Colombia · Universidad Autónoma de Occidente GobLab-UAI Cali, Colombia
Beyond wheelchairs and blindfolds: Investigating disability stereotypes in T2I models with INCLUDE-BENCH
Jul 9, 2026Sophia Lichtenberg, Albert Gatt, Judith MasthoffDisability Modern Text-To-Image Models
Utrecht University
When Debiasing Backfires: Counterintuitive Side Effects of Preprocessing-Based Stereotype Mitigation
Jul 8, 2026Yahan Zheng, John Guerrerio, Soroush Vosoughi +1Stereotype Mitigation Preprocessing
Dartmouth College · Oakland University
Scalable and Culturally Specific Stereotype Dataset Construction via Human-LLM Collaboration
Jul 8, 2026Weicheng Ma, John Guerrerio, Soroush VosoughiStereotype Mitigation Multilingual Benchmark
College of Computing, Georgia Institute of Technology · Computer Science Department, Dartmouth College
Training-Free Debiasing of Diffusion Models via CLIP-Guided Denoising Optimization
Jul 1, 2026Dain Kim, Jinseo Kim, Sungyong BaikText-To-Image Diffusion Models Diffusion Models
Dept. of Artificial Intelligence, Hanyang University, South Korea · Dept. of Data Science, Hanyang University, South Korea
Wait, am I Being Fair? Characterizing Deductive Stereotyping and Mitigating It with Fair-GCG
Jun 30, 2026Naihao Deng, Yilun Zhu, Joan Nwatu +2Stereotype Mitigation Algorithmic Fairness
University of Michigan
Unmasking LAION-5B: Age, Gender, Race, and Emotion Biases in Large-Scale Image Datasets
Jun 22, 2026Iris Dominguez-Catena, Daniel Paternain, Mikel GalarDemographics Large-Scale Image Datasets
Institute of Smart Cities (ISC), Department of Statistics, Computer Science and Mathematics · Public University of Navarre · Arrosadia Campus +1
The Wrong Kind of Right: Quantifying and Localizing Misfired Alignment in LLMs
Jun 17, 2026Naihao Deng, Yiming Feng, Chimaobi Okite +4Large Language Model Alignment Misalignment Persona
University of Michigan · University of Cambridge · University of Aberdeen
AmchiBias: Measuring Stereotypical Bias in Goan Identity Groups with a Minimal Pair Dataset in English and Konkani
Jun 13, 2026Michelle Barbosa, Sebastian Padó, Franziska WeeberStereotype Mitigation Language Modeling
Institute for Natural Language Processing, University of Stuttgart
Characterizing Cultural Localization in AI-Generated Stories
Jun 12, 2026Shaily Bhatt, Supriti Vijay, Jeremiah Milbauer +1Narrative Generation Narratives
Carnegie Mellon University
It Takes One to Bias Them All: Breaking Bad with One-Shot GRPO
Jun 9, 2026Naihao Deng, Yilun Zhu, Naichen Shi +2Large Language Model Bias Large Language Model Alignment
University of Michigan · Northwestern University
Neutrality Bites: Gender Representation in AI-Generated Animal Stories
Jun 6, 2026Imani Finkley, Yuanxi Li, Melanie WalshGender Bias Gender
University of Washington, Seattle, Washington, USA
Identifying High-Confidence Social Biases in LLMs for Trustworthy Conversational Tutoring Agents
Jun 1, 2026Aitor Arronte Alvarez, Naiyi Xie FinchamLarge Language Model Bias Tutors
University of Hawaii at Manoa, Honolulu HI 96822, USA
Anchoring LLM Gender Bias to Human Baselines: A Cross-Lingual Audit
May 29, 2026Jiwoo Choi, Seonwoo Ahn, Tongxin Zhang +1Large Language Model Bias Stereotype Mitigation
School of Digital Humanities and Computational Social Sciences, KAIST, South Korea
GRUFF: LLM Pronoun Fidelity, Reasoning, and Biases in German
May 28, 2026Fabian Mewes, Anne Lauscher, Vagrant GautamLarge Language Model Bias Gender
JobMatchMe GmbH, Germany · Trustworthy AI Lab, University of Hamburg, Germany · Heidelberg Institute for Theoretical Studies, Germany
Side-by-side Comparison Amplifies Dialect Bias in Language Models
May 23, 2026Kritee Kondapally, Claire J. Smerdon, Pooja C. Patel +5Large Language Model Bias Dialects
University of Southern California, USA
Assisted Counterspeech Writing at the Crossroads of Hate Speech and Misinformation
May 21, 2026Genoveffa Martone, Helena Bonaldi, Marco GueriniHate Speech Misinformation
Fondazione Bruno Kessler, Italy · Università Cattolica del Sacro Cuore, Italy
StereoTales: A Multilingual Framework for Open-Ended Stereotype Discovery in LLMs
May 11, 2026Pierre Le Jeune, Étienne Duchesne, Weixuan Xiao +4Stereotype Mitigation Biases
Giskard AI · École Normale Supérieure, INSERM, Paris
Progressing beyond Art Masterpieces or Touristic Clichés: how to assess your LLMs for cultural alignment?
Apr 28, 2026António Branco, João Silva, Nuno Marques +10Cultural Awareness Large Language Model Alignment
†University of Lisbon · ‡University of Beira Interior
Intersectional Fairness in Large Language Models
Apr 22, 2026Chaima Boufaied, Ronnie De Souza Santos, Ann BarcombLarge Language Model Bias Stereotype Mitigation
University of Calgary, Calgary, Canada
Cross-Cultural Value Attribution in Large Vision-Language Models
Apr 10, 2026Phillip Howard, Xin Su, Kathleen C. FraserStereotype Mitigation Moral Reasoning
Thoughtworks · University of Ottawa
Evaluating Implicit Biases in LLM Reasoning through Logic Grid Puzzles
Nov 8, 2025Fatima Jahara, Mark Dredze, Sharon LevyLarge Language Model Bias LLM Reasoning Strategies
Rutgers University · Johns Hopkins University
Fair-GPTQ: Bias-Aware Quantization for Large Language Models
Sep 18, 2025Irina Proskurina, Guillaume Metzler, Julien VelcinLarge Language Model Quantization Algorithmic Fairness
Université Claude Bernard Lyon 1, Université Lumière Lyon 2, ERIC · École Centrale de Lyon, LIRIS, CNRS UMR 5205
BiasGym: A Simple and Generalizable Framework for Analyzing and Removing Biases through Injection
Aug 12, 2025Sekh Mainul Islam, Nadav Borenstein, Siddhesh Milind Pawar +3Large Language Model Bias Biases
University of Copenhagen
AI's Blind Spots: Geographic Knowledge and Diversity Deficit in Generated Urban Scenario
Jun 20, 2025Ciro Beneduce, Massimiliano Luca, Bruno LepriText-To-Image Diffusion Models Diffusion Models
* Bruno Kessler Foundation · **University of Trento
When majority rules, minority loses: bias amplification of gradient descent
May 19, 2025François Bachoc, Jérôme Bolte, Ryan Boustany +1Minority Imbalanced Classification
LPP · University of Lille · Institut Universitaire de France (IUF) +6
General Phrase Debiaser: Debiasing Masked Language Models at a Multi-Token Level
Nov 23, 2023Bingkang Shi, Xiaodan Zhang, Dehan Kong +4Debiasing Large Language Model Bias
Chinese Academy of Sciences, Institute of Information Engineering, Beijing 100093, China · Alibaba Group, Alibaba Artificial Intelligence Governance Laboratory, Beijing, China · University of Bristol, Department of Electrical and Electronic Engineering, Bristol, BS8 1UB, UK

Oct 1, 2026, cs.CLJ/K move · Enter open · S save

Acmite: Mitigating Gender Bias in LLMs through Concept-Guided Mutual Information

Tian Lan, Xiaoqing Cheng, Han Zhang, Jiang Li

Kyoto University · Zhejiang University · Shanghai Jiao Tong University+1

Gender Bias Stereotype Mitigation Debiasing Mutual Information

Open paper AskPDF arXiv

Large language models (LLMs) can reproduce social stereotypes from their training data, motivating extensive research on model debiasing. However, existing methods often rely on explicit biased examples or predefined group-term substitutions, making them sensitive to wording and less effective at capturing stereotype concepts shared across diverse contexts. More importantly, they typically suppress biased outputs without explicitly modeling the statistical dependence between model outputs and the underlying stereotype concepts. We propose Acmite, a lightweight concept-guided framework for targeted and selective debiasing. Acmite represents stereotypes as structured semantic concepts and uses maximal marginal relevance (MMR) to select diverse concepts for debiasing. Inspired by mutual information minimization, it approximates this dependence with token-level KL divergence while preserving task semantics. A lightweight LoRA adapter is trained with the base model frozen and activated at inference time only when the input is sufficiently similar to stereotype-related concepts; otherwise, the original model is used directly. We evaluate Acmite on BBQ, CrowS-Pairs, and StereoSet, and assess general capability preservation on ARC-Challenge, GSM8K, and PIQA. Experiments across three LLMs show that Acmite effectively mitigates gender bias across complementary evaluation formats while maintaining competitive performance on bias-unrelated tasks. Anonymous code and data are available at https://anonymous.4open.science/r/Acmite-18E2/.

Momentum

Related topics

Latest papers 42