Social Bias in Language Models

Recent advances in multilingual representation learning aim to bridge the performance gap between high- and low-resource languages, yet their ability to preserve affective meaning across languages remains underexplored, particularly for underrepresented languages like Bengali. This research addresses cross-lingual sentiment misalignment between Bengali and English by introducing a controlled benchmarking framework evaluating four multilingual transformer models on parallel Bengali-English sentence pairs, stratified by dialect, to assess their representational stability. We demonstrate that a compressed model architecture exhibits a 28.7% "Sentiment Inversion Rate," fundamentally misinterpreting positive semantics as negative (or vice versa). Consequently, we identify a cross-lingual sentiment skew that we call "Asymmetric Empathy," where models systematically dampen or artificially amplify the affective weight of Bengali text relative to its exact English counterpart. Finally, we expose a key vulnerability regarding dialectal representation: a "Modern Bias" in the regional model, which exhibits a 57% increase in alignment error when processing the formal Bengali register compared to modern colloquial text. As foundational encoders continue to serve as safety classifiers and reward models for LLM pipelines, cross-lingual reliability becomes a critical concern. We therefore advocate for the integration of "Affective Stability" metrics into future cross-lingual benchmarks to detect and penalize polarity inversions, particularly in low-resource settings.

Large Language Models (LLMs) are increasingly deployed in contact-center Quality Assurance (QA) to automate agent performance evaluation and coaching feedback. While LLMs offer unprecedented scalability and speed, their reliance on web-scale training data raises concerns regarding demographic and behavioral biases that may distort workforce assessment. We present a counterfactual fairness evaluation of LLM-based QA systems across 13 dimensions spanning three categories: Identity, Context, and Behavioral Style. Fairness is quantified using the Counterfactual Flip Rate (CFR), the frequency of binary judgment reversals, and the Mean Absolute Score Difference (MASD), the average shift in coaching or confidence scores across counterfactual pairs. Evaluating 18 LLMs on 3,000 real-world contact center transcripts, we find systematic disparities, with CFR ranging from 5.4% to 13.0% and consistent MASD shifts across confidence, positive, and improvement scores. Larger, more strongly aligned models show lower unfairness, though fairness does not track accuracy. Contextual priming of historical performance induces the most severe degradations (CFR up to 16.4%), while implicit linguistic identity cues remain a persistent bias source. Finally, we analyze the efficacy of fairness-aware prompting, finding that explicit instructions yield only modest improvements in evaluative consistency. Our findings underscore the need for standardized fairness auditing pipelines prior to deploying LLMs in high-stakes workforce evaluation.

With natural language generation becoming a popular use case for language models, the Bias Benchmark for Question-Answering (BBQ) has grown to be an important benchmark format for evaluating stereotypical associations exhibited by generative models. We expand the linguistic scope of BBQ and construct FilBBQ through a four-phase development process consisting of template categorization, culturally aware translation, new template construction, and prompt generation. These processes resulted in a bias test composed of more than 10,000 prompts which assess whether models demonstrate sexist and homophobic prejudices relevant to the Philippine context. We then apply FilBBQ on models trained in Filipino but do so with a robust evaluation protocol that improves upon the reliability and accuracy of previous BBQ implementations. Specifically, we account for models' response instability by obtaining prompt responses across multiple seeds and averaging the bias scores calculated from these distinctly seeded runs. Our results confirm both the variability of bias scores across different seeds and the presence of sexist and homophobic biases relating to emotion, domesticity, stereotyped queer interests, and polygamy. FilBBQ is available via https://github.com/gamboalance/filbbq.

Post-training quantization reduces the memory needed to run large language models but alters their social biases in ways that aggregate metrics fail to capture. We present the first large-scale study of 50 quantized models evaluated on PostTrainingBiasBench, a unified benchmark of 13 closed- and open-ended bias datasets. We identify a phenomenon we term quantization-induced bias flipping, in which quantization causes models to change responses from biased to unbiased and vice versa, up to 21% of the time, despite no change in aggregate bias scores. These flips are strongly associated with model uncertainty, where the responses with high uncertainty are 3-11x more likely to change than the confident ones. Quantization strength amplifies this effect, with 4-bit quantized models exhibiting 4-6x more behavioral changes than 8-bit quantized models. Critically, these changes create asymmetric impacts across demographic groups, where bias can worsen by up to 18.6% for some groups while improving by 14.1% for others, yielding misleadingly neutral aggregate outcomes. Larger models show no consistent robustness advantage, and group-specific shifts vary unpredictably across model families. Our findings demonstrate that compression fundamentally alters bias patterns, requiring crucial post-quantization evaluation and interventions to ensure reliability in practice.

Large language models (LLMs) exhibit social biases that reinforce harmful stereotypes, limiting their safe deployment. Most existing debiasing methods adopt a suppressive paradigm by modifying parameters, prompts, or neurons associated with biased behavior; however, such approaches are often brittle, weakly generalizable, data-inefficient, and prone to degrading general capability. We propose KnowBias, a lightweight and conceptually distinct framework that mitigates bias by strengthening, rather than suppressing, neurons encoding bias-knowledge. KnowBias identifies neurons encoding bias knowledge using a small set of bias-knowledge questions via attribution-based analysis, and selectively enhances them at inference time. This design enables strong debiasing while preserving general capabilities, generalizes across bias types and demographics, and is highly data efficient, requiring only a handful of simple yes/no questions and no retraining. Experiments across multiple benchmarks and LLMs demonstrate consistent state-of-the-art debiasing performance with minimal utility degradation. Data and code are available at https://github.com/JP-25/KnowBias.

Algorithmic audits are essential tools for examining systems for properties required by regulators or desired by operators. Current audits of large language models (LLMs) primarily rely on black-box evaluations that assess model behavior only through input-output testing. These methods are limited to tests constructed in the input space, often generated by heuristics. In addition, many socially relevant model properties (e.g., gender bias) are abstract and difficult to measure through text-based inputs alone. To address these limitations, we propose a white-box sensitivity auditing framework for LLMs that leverages activation steering to conduct more rigorous assessments through model internals. Our auditing method conducts internal sensitivity tests by manipulating key concepts relevant to the model's intended function for the task. We demonstrate its application to bias audits in four simulated high-stakes LLM decision tasks. Our method consistently indicates substantial dependence on protected attributes in model predictions, even in settings where standard black-box evaluations suggest little or no bias. Our code is openly available at https://github.com/hannahxchen/llm-steering-audit

As large language models (LLMs) are adopted into frameworks that grant them the capacity to make real decisions, it is increasingly important to ensure that they are unbiased. In this paper, we argue that the predominant approach of simply removing existing biases from models is not enough. Using a paradigm from the psychology literature, we demonstrate that LLMs can spontaneously develop novel social biases about artificial demographic groups even when no inherent differences exist. These biases result in highly stratified task allocations, which are less fair than assignments by human participants and are exacerbated in newer and larger models. In humans, emergent biases like these have been shown to result from exploration-exploitation trade-offs, where the decision-maker explores too little, allowing early observations to strongly influence impressions about entire demographic groups. To alleviate this effect, we explore a series of interventions targeting model inputs, problem structure, and explicit steering. While most interventions have limited effect, explicitly incentivizing exploration robustly reduces stratification, highlighting the need for better multifaceted objectives to mitigate bias. These results reveal that LLMs are not merely passive mirrors of human social biases, but can actively create new ones from experience, raising urgent questions about how these systems will shape societies over time.

Large language models (LLMs) are now widely deployed in user-facing applications, reaching hundreds of millions of users worldwide. Despite their widespread adoption, growing reliance on their outputs raises significant concerns, particularly as users may be exposed to model-inherent biases that disadvantage or stereotype certain groups. However, existing bias benchmarks commonly rely on simple templated prompts or restrictive multiple-choice questions that fail to capture the complexity of real-world user interactions. In this work, we address this gap by introducing a counterfactual framework that automatically generates realistic, open-ended questions for LLM bias evaluation. Through iterative question mutation, our approach systematically explores areas where models are most likely to exhibit biased behavior. Beyond just detecting harmful biases, we also capture increasingly relevant response dimensions, such as asymmetric refusals and explicit bias acknowledgment. Building on this, we construct CAB, a diverse and human-verified benchmark for realistic and nuanced bias evaluations on current frontier LLMs. Our evaluation using CAB highlights the continued need for fairness research by showing that all examined models exhibit persistent biases across certain scenarios.

Large language models (LLMs) are increasingly under scrutiny for perpetuating identity-based discrimination in high-stakes domains such as hiring, particularly against people with disabilities (PwD). However, existing research remains largely Western-centric, overlooking how intersecting forms of marginalization--such as gender and caste--shape experiences of PwD in the Global South. We conduct a comprehensive audit of six LLMs across 2,820 hiring scenarios spanning diverse disability, gender, nationality, and caste profiles. To capture subtle intersectional harms and biases, we introduce ABLEIST (Ableism, Inspiration, Superhumanization, and Tokenism), a set of five ableism-specific and three intersectional harm metrics grounded in disability studies literature. Our results reveal significant increases in ABLEIST harms towards disabled candidates--harms that many state-of-the-art models failed to detect. These harms were further amplified by sharp increases in intersectional harms (e.g., Tokenism) for gender and caste-marginalized disabled candidates, highlighting critical blind spots in current safety tools and the need for intersectional safety evaluations of frontier models in high-stakes domains like hiring.

The high memory demands of generative language models have drawn attention to quantization, which reduces memory usage by mapping model weights to lower-precision integers. However, recent empirical studies show that, while efficient, quantization can increase the likelihood of generating biased outputs and degrade performance on fairness benchmarks. In this work, we draw new links between quantization and model fairness by adding explicit group-fairness constraints to the quantization objective and introduce Fair-GPTQ, the first quantization method explicitly designed to reduce unfairness in large language models. The added constraints guide the learning of the rounding operation toward less-biased text generation for protected groups. Specifically, we focus on stereotype generation involving occupational bias and discriminatory language spanning gender, race, and religion. Fair-GPTQ has minimal impact on performance, preserving at least 90% of baseline accuracy on zero-shot benchmarks, reduces unfairness relative to a half-precision model, and retains the memory and speed benefits of 4-bit quantization.

Understanding biases and stereotypes encoded in the weights of Large Language Models (LLMs) is crucial for developing effective mitigation strategies. However, biased behavior is often subtle and non-trivial to isolate, even when deliberately elicited, making systematic analysis and debiasing particularly challenging. To address this, we introduce a simple, cost-effective, and generalizable framework BiasGym for reliably injecting, analyzing, and mitigating conceptual associations of biases within LLMs. BiasGym consists of two modules: Inject, which injects specific biases into the model via token-based fine-tuning while keeping the model frozen, followed by two debiasing methods that leverage these injected signals to identify and reliably suppress (Scope) or Steer the components responsible for biased behavior. Our framework enables consistent bias elicitation for better localization of bias conceptual association in the model space, supports targeted debiasing without degrading performance on downstream tasks, and generalizes to biases unseen during fine-tuning. We demonstrate the effectiveness of our proposed framework in reducing real-world stereotypes (e.g., people from Italy being `reckless drivers'), showing its utility for both safety interventions and interpretability research.

One way social groups are marginalized in discourse is that the narratives told about them often default to a narrow, stereotyped range of topics. In contrast, default groups are allowed the full complexity of human existence. We describe the constrained representations of queer people in LLM generations in terms of harmful representations, narrow representations, and discursive othering and formulate hypotheses to test for these phenomena. Our results show that LLMs are significantly limited in their portrayals of queer personas.

Ensuring the trustworthiness of Multimodal Large Language Models (MLLMs) requires rigorous evaluation of gender bias, particularly in socially sensitive domains. Existing benchmarks primarily assess bias in isolated contexts and are less suited to capturing biases that emerge through interpersonal interactions. We introduce Genres, a benchmark for evaluating gender bias in MLLMs through the lens of social relationships in narrative generation. Grounded in Fiske's relational models theory, Genres comprises 1,440 multimodal narrative elicitation pairs spanning four fundamental relationship types, diverse social scenarios, and three age groups. Its dual-character generation tasks elicit socially situated interpersonal narratives and enable fine-grained assessment across multiple dimensions of gender bias. Experiments on six recent MLLMs reveal persistent yet strongly relationship- and model-dependent gender disparities, with some patterns becoming more pronounced in interpersonal settings. We further evaluate two complementary generation-stage mitigation strategies. Our findings demonstrate the value of relationship-aware evaluation for uncovering subtle, interaction-driven gender bias in MLLMs.

Large language models (LLMs) are increasingly used in high-stakes decisions such as hiring and college admissions, making their social bias a critical concern. While LLMs are trained to refuse explicitly biased requests, bias can be leaked implicitly during LLM planning and reasoning process. As code becomes the primary medium for LLM internal logic-writing, we introduce FairCoder, a benchmark that frames decision-making as coding tasks to systematically probe LLM bias across employment, education, and healthcare domains, covering multiple fairness definitions. Considering that existing metrics may fail when LLMs frequently refuse the request, we propose FairScore, a metric that jointly captures refusal behavior and group-level outcome diversity. Experiments with a 1k-sample dataset on powerful LLMs reveal consistent and previously underexplored bias patterns, such as prioritizing applicants from high-income families in college admissions. Our findings highlight the risks of deploying LLMs as decision-making agents and provide a comprehensive evaluation framework for future research.

Large language models (LLMs) reproduce homogeneity bias -- the tendency to portray marginalized groups as more internally similar than dominant groups -- but whether this bias is stable or an artifact of inference settings has only been studied in single proprietary models. We map homogeneity bias across a 5x5 temperature-by-top-p grid in seven open-weight instruction-tuned LLMs (7-20B parameters). Hispanic and Asian Americans are portrayed as more homogeneous than White Americans in at least 18 of 20 hyperparameter configurations across six of seven models, including at extreme sampling settings. African American and gender bias show model-specific variation in direction. A conservative cell-level re-analysis confirms Hispanic and Asian homogeneity as robust, while weaker African American and gender signals largely do not survive, establishing group-specific robustness. We also apply the same grid to a names-based paradigm in which group identity is signaled via racially distinctive surnames rather than explicit labels. The names paradigm corroborates Hispanic and Asian homogeneity bias, but Black-coded surnames elicit robustly less homogeneous outputs than White-coded names in every model tested -- a reversal absent from the label paradigm -- showing that how group identity is operationalized shapes which biases surface and in which direction.

Large language models (LLMs) may encode biased associations from heterogeneous training corpora that are not immediately visible under ordinary prompting, but can surface when the model is steered toward particular demographic personas. Such behavior often manifests not as explicit toxic output, but as systematic performance differences across semantically equivalent tasks, making the resulting bias difficult to detect and mitigate. To address this issue, we formalize the implicit bias problem as persona-induced performance disparity and argue that bias evidence should be treated as a graded signal rather than a binary label. Motivated by this observation, we model biased knowledge as a fuzzy subset equipped with an explicit membership function that reflects the strength of bias evidence for each candidate example. Building on this formulation, we propose Bayesian-Theory-based Bias Removal (BTBR), a hybrid probabilistic-fuzzy framework for identifying and removing latent bias traces from model parameters. BTBR first performs likelihood-ratio screening to measure how strongly candidate samples align with a target biased persona, then converts high-membership samples into structured knowledge triples, and finally applies targeted model editing with a lightweight fuzzy rule scheduler to reduce collateral performance degradation under high entanglement risk. Extensive experiments across multiple bias sources, tasks, model families and editing backends show that BTBR consistently reduces persona-induced performance gaps while preserving general reasoning ability. These results demonstrate that combining probabilistic evidence with fuzzy degree modeling provides an effective and practical approach for mitigating implicit bias in large language models.

Parameter-efficient fine-tuning (PEFT) has become a de facto standard for adapting large language models (LLMs). However, we identify a critical vulnerability within popular low-rank adaptation methods such as LoRA: they can exacerbate "Catastrophic Inheritance" - the unchecked propagation of biases, noise, and data imbalances from pre-training. This phenomenon can degrade model robustness and fairness, undermining the benefits of efficient adaptation. To address this, we introduce Bias-Alleviating Low-Rank Adaptation (BA-LoRA). Our approach is founded on a principled decomposition of Catastrophic Inheritance into three core challenges: Knowledge Drift, Representation Collapse, and Overfitting to Noise. BA-LoRA systematically mitigates these issues by incorporating a trio of targeted regularizers: consistency, diversity, and an SVD-based term, designed to preserve core knowledge, promote representational richness, and encourage robust, low-rank output representations, respectively. We conduct comprehensive evaluations on a suite of Natural Language Generation (NLG) and Natural Language Understanding (NLU) tasks using diverse, prominent open-source language models (e.g., LLaMA-2-7B and DeBERTa-v3-base). Our results show that BA-LoRA not only outperforms state-of-the-art LoRA variants in terms of performance and stability, but also demonstrates superior robustness and bias mitigation on targeted evaluations. These results provide evidence that BA-LoRA can counteract the adverse effects of Catastrophic Inheritance.

Language Models (LMs) have demonstrated exceptional performance across various Natural Language Processing (NLP) tasks. Despite these advancements, LMs can inherit and amplify societal biases related to sensitive attributes such as gender and race, limiting their adoption in real-world applications. Therefore, fairness has been extensively explored in LMs, leading to the proposal of various fairness notions. However, the lack of clear agreement on which fairness definition to apply in specific contexts and the complexity of understanding the distinctions between these definitions can create confusion and impede further progress. To this end, this paper proposes a systematic survey that clarifies the definitions of fairness as they apply to LMs. Specifically, we begin with a brief introduction to LMs and fairness in LMs, followed by a comprehensive, up-to-date overview of existing fairness notions in LMs and the introduction of a novel taxonomy that categorizes these concepts based on their transformer architecture: encoder-only, decoder-only, and encoder-decoder LMs. We further illustrate each definition through experiments, showcasing their practical implications and outcomes. Finally, we discuss current research challenges and open questions, aiming to foster innovative ideas and advance the field. The repository is publicly available online at https://github.com/vanbanTruong/Fairness-in-Large-Language-Models/tree/main/definitions.

Generative large language models (LLMs) have been shown to exhibit harmful biases and stereotypes. While safety fine-tuning typically takes place in English, if at all, these models are being used by speakers of many different languages. There is existing evidence that the performance of these models is inconsistent across languages and that they discriminate based on demographic factors of the user. Motivated by this, we investigate whether the social stereotypes exhibited by LLMs differ as a function of the language used to prompt them, while controlling for cultural differences and task accuracy. To this end, we present MBBQ (Multilingual Bias Benchmark for Question-answering), a carefully curated version of the English BBQ dataset extended to Dutch, Spanish, and Turkish, which measures stereotypes commonly held across these languages. We further complement MBBQ with a parallel control dataset to measure task performance on the question-answering task independently of bias. Our results based on several open-source and proprietary LLMs confirm that some non-English languages suffer from bias more than English, even when controlling for cultural shifts. Moreover, we observe significant cross-lingual differences in bias behaviour for all except the most accurate models. With the release of MBBQ, we hope to encourage further research on bias in multilingual settings. The dataset and code are available at https://github.com/Veranep/MBBQ.

The social biases and unwelcome stereotypes revealed by pretrained language models are becoming obstacles to their application. Compared to numerous debiasing methods targeting word level, there has been relatively less attention on biases present at phrase level, limiting the performance of debiasing in discipline domains. In this paper, we propose an automatic multi-token debiasing pipeline called General Phrase Debiaser, which is capable of mitigating phrase-level biases in masked language models. Specifically, our method consists of a phrase filter stage that generates stereotypical phrases from Wikipedia pages as well as a model debias stage that can debias models at the multi-token level to tackle bias challenges on phrases. The latter searches for prompts that trigger model's bias, and then uses them for debiasing. State-of-the-art results on standard datasets and metrics show that our approach can significantly reduce gender biases on both career and multiple disciplines, across models with varying parameter sizes.