Bias & fairness

This paper develops a variational framework for regulated language generation. Starting from autoregressive token sampling, we derive the induced distribution over complete messages and relate it to an entropy-regularized Gibbs law. Regulation is modeled as an optimal discriminator whose convex-dual value is an f-divergence, and the generator-regulator interaction is formulated as a saddle-point problem. The framework applies to moderation, censorship, AI deception detection, compliance auditing

Fairness is an important aspect of decision-making in multi-objective reinforcement learning (MORL), where policies must ensure both optimality and equity across multiple, potentially conflicting objectives. While single-policy MORL methods can learn fair policies for fixed user preferences using welfare functions such as the generalized Gini welfare function (GGF), they fail to provide the diverse set of policies necessary for dynamic or unknown user preferences. To address this limitation, we

In this paper, we establish a set of theoretical impossibility results, termed the No-Free-Fairness theorems, that identify three fundamental sources of disparity in learning systems. First, we show that when a task exhibits irreducible cost on a subgroup, any decision rule must trade off overall performance with disparity, yielding an inherent fairness--cost frontier. Second, we prove that even in ideal, noise-free settings where a perfectly fair and accurate solution exists, finite-sample lear

Graph-based learning methods have become increasingly prominent due to their strong performance across diverse applications. Among these, recent frameworks grounded in diffusion processes provide a unifying perspective that extends traditional graph neural network formulations while addressing limitations of standard message-passing mechanisms. Despite these advances, concerns remain regarding the fairness of such models, as they may propagate or amplify biases present in the data. In this work,

Large Language Model (LLM) unlearning aims to remove undesirable knowledge or behaviors while preserving retained capabilities. Current unlearning methods all involve a trade-off between unlearning and retention. We have found that the retention activation bias can also be used to quantify the damage an unlearning method inflicts on retention, without considering the specific implementation of the unlearning process. This allows us to restore retention performance for any unlearning method using

Existing vision-language model (VLM)-based AI-generated image quality assessment (AIGIQA) methods suffer from a fundamental semantic-distortion dimensional conflict: monolithic representations optimized for semantic discrimination inherently entangle compositional understanding with low-level perceptual sensitivity, rendering them blind to fine-grained quality degradations. We introduce MST-CLIPIQA, a multi-scale two-stream framework that achieves hierarchical vision-language alignment through e

Large language model (LLM) agents increasingly take actions (screening applicants, recommending credit, triaging patients), yet fairness for LLMs is still measured by grading answers. We introduce AgentFairBench, a cheap, reproducible, multi-domain benchmark for demographic disparity in the actions of LLM agents. Grounded in a companion framework, the Bias Conduction Framework (BCF, restated here), it spans three regulator-anchored domains: hiring, lending, and medical triage. Synthetic, demogra

Based on a questionnaire of 100 higher-education students, predominantly from engineering-related fields, and a critical review of recent literature, this chapter examines how students use and perceive Large Language Models (LLMs) in engineering education. Students primarily value LLMs for writing support, conceptual clarification, coding assistance, and brainstorming, while simultaneously expressing concerns about inaccuracies, bias, overreliance, academic integrity, and the burden of verificat

Entity-aware document retrieval uses query-associated entities as ranking signals, assuming that semantically relevant entities are also useful retrieval signals. We show this assumption is insufficient- and explain why. Unlike terms, which are ground-truth observations, entity links are hypotheses produced by an imperfect linker: an entity can be topically central yet provide no discriminative signal if the linker fires indiscriminately across relevant and non-relevant documents. We formalize t

Visual perception of urban streetscapes underpins evidence-based decisions in landscape planning, public health, and place-making. Yet models trained on a few well-photographed metropolises systematically misjudge underrepresented districts, propagating geographic bias into downstream policy. We address this gap with HVSP-LL, a lifelong learning framework that couples a stratified visual-semantic pivoting module with an equity-aware rehearsal mechanism. The pivoting module organises landscape co

The purpose of this qualitative multi-case study was to examine how social bias emerges, is perceived, and can be mitigated within artificial intelligence and machine learning systems by practitioners directly involved in their design, development, and governance. Although examples from healthcare, criminal justice, employment, and education were used to illustrate domains where automated systems shape everyday life, the study focused on the lived experiences and professional insights of AI prac

This work introduces the Separable Neural Architecture (SNA), a function representational class combining neural approximation with tensor decomposition. The SNA decouples localized coordinate functions (atoms) from global interactions governed by a sparse, low-rank interaction object. This architecture possesses a compact and smooth inductive bias well-suited for solving partial differential equations (PDEs). When viewed as a Galerkin trial space under the variational SNA (VSNA) framework, the

MLLM-as-a-Judge is conventionally validated by agreement with human annotations, but this metric is undefined when the human pool is culturally heterogeneous. We introduce VOIR DIRE, a multimodal benchmark of 626 culturally paired image--prompt artifacts spanning U.S. and mainland Chinese contexts across food, fashion, and architecture, with annotator pools that are within-pool reliable (a = 0.86/0.74) but cross-pool divergent on evaluation (Q1 r = -0.12). Across six MLLMs, the bias decomposes i

The proliferation of recursive training on synthetic data can alleviate data scarcity but risks model collapse, where repeated training erodes distributional tails and homogenizes outputs. Data selection is widely viewed as a remedy, yet its reliability depends critically on the reference distribution used by the verifier. We show that in low-resource verification regimes, where each verifier observes only a small, fragmented, and biased slice of the target manifold, selection itself becomes bia

Interfacing Artificial Intelligence (AI) with democracy is one of the most profound challenges of our times. On the one hand, AI comes with opportunities to overcome long-standing challenges in democracy, such as low participation in deliberative and voting processes with poor representation of people. On the other hand, new risks arise from AI algorithms that are privacy-intrusive, biased, manipulative, spread misinformation and influence election results. Moving beyond the over-simplistic ques

Correcting handwritten exams by hand is time-consuming and error-prone, particularly for large cohorts, while fully digital exams tend to force a didactic narrowing towards closed question formats. A practical middle ground keeps paper-based, problem-oriented tasks but records the assessment-relevant answers as single capital letters in a table that a machine can read. The open question is whether this reading can be made accurate and, above all, fair enough for unsupervised grading. Earlier aut

Behavioral biometrics offer a promising approach for continuous authentication, but their fairness across demographic groups remains largely unexplored. This paper investigates gender bias in swipe-based authentication using the BBMAS (117 users) and ANTAL (71 users) datasets and evaluates XGBoost and DenseNet classifiers through False Acceptance Rate (FAR) and False Rejection Rate (FRR). XGBoost achieved authentication accuracies of 92% and 94% on the BBMAS and ANTAL datasets, respectively, whi

Claims about the robustness and fairness of deepfake speech detectors are only as credible as the datasets used to train and evaluate those systems. We present a dataset-level audit of the deepfake speech landscape. We compile and analyze 39 deepfake speech datasets, examining key attributes including accessibility, documentation, demographic and language coverage, dataset scale, and the underlying bona fide speech sources. Our audit reveals two important takeaways. Firstly, fairness assessment

Social media has become a key channel for political advertising during election campaigns. However, algorithmic biases in the delivery of these ads may distort the public's exposure to political messaging. This can hinder citizens' ability to make informed choices and undermine equal access to political discourse, raising concerns about the integrity of electoral processes. In this study, we examine gender-based discrimination in the delivery of political ads during the 2024 European Parliament

Lipschitz-style individual fairness formalizes the idea that semantically similar examples should receive similar predictions, but its evaluation in multi-task learning (MTL) can be confounded by method-induced representation scales. This paper identifies threshold confounding: when the auditing tolerance is derived from each model's own representation distances, different algorithms are compared under different semantic thresholds. A threshold-drift analysis further shows how Bias rankings can

Adversarial examples reveal vulnerabilities in Vision-Language Pre-training (VLP) models and provide insights for improving robustness. A key property is cross-model transferability, which enables transfer-based black-box attacks. However, existing attacks often rely heavily on the surrogate model, causing cross-model performance drops. One reason is that adversarial optimization may follow surrogate model responses more than input semantics, making the update direction effective on the surrogat

Personalized persuasive text generation can improve relevance and engagement, but demographic conditioning may also introduce unequal framing across groups. We study fairness mitigation in personalized generation as a constrained multi-objective alignment problem: reduce demographic disparities while preserving personalization fidelity. We propose a Pareto-guided teacher alignment framework that combines revision-based candidate generation, pair-aware feasibility gating, Pareto-style candidate s

In high-stakes settings such as brand compliance, clinical care, and content moderation, machine learning cannot be deployed as opaque oracles: practitioners inspect the features driving model decisions, and models must leverage the expert documentation governing these domains. In practice, the data arrives as unstructured content, and features extracted from it must be interpretable, discriminative, and aligned with what experts consider important. Existing methods fall short: they target tabul

We present an online reinforcement learning (RL) algorithm for fine-tuning flow-matching policies in continuous-control problems. Our key insight is to view RL-based policy improvement as a transport of action densities towards regions of high reward, which naturally aligns with the transport formulation of flow matching models. Prior methods either approximate the current or optimal policy distribution or resort to distillation, which introduces biased gradients or sacrifices multimodal modelin

Background: Consumer-facing large language models are now a common source of health information, and they interpret and personalize responses rather than retrieve them. Whether their responses vary across users is a clinical, equity, and governance question, sharpened by evidence that sycophantic responses can alter judgment and increase trust. Objective: To evaluate response variation and sycophancy in consumer-facing health LLMs under conditions resembling ordinary patient use. Methods: We con

Graduate-level research reading report assessment creates a substantial labor burden for educators. While large language models (LLMs) hold great potential for automating academic grading, their reliability for this specialized task remains understudied, particularly regarding grading consistency, the lack of which represents a primary obstacle to educational fairness. This paper proposes a human-aligned LLM-assisted grading workflow and presents a case study based on 180 student submissions fro

Predicting a user's next Point-of-Interest (POI) based on their historical check-in records is a fundamental task in location-based services. While recent methods incorporating large language models have shown strong reasoning capabilities and promising results, they typically formulate the prediction task as a one-step trajectory-to-location mapping problem, making predictions prone to shallow trajectory correlations and historical frequency bias. We argue that users rarely choose locations dir

Pretraining is fundamental to the development of Large Language Models (LLMs), yet the opacity of pretraining data complicates model analysis and raises ethical, legal, and fairness concerns. Detecting whether specific datasets were used during pretraining is, therefore, critical. Existing state-of-the-art methods typically rely on access to model probability distributions, making them unsuitable for closed-source LLMs that provide only input-output interfaces. To address this limitation, we int

Large language models (LLMs) increasingly rely on reward models to align their outputs with diverse user preferences. While personalized reward models aim to capture such heterogeneity, they are often trained on imbalanced user preference data and may therefore favor users whose preferences are more common in the training population. In this paper, we identify this failure mode as personalized reward bias, where reward modeling quality varies systematically with preference support rate. We formu

Gender bias in AI-generated stories is a well-documented problem. While much attention has been paid to reducing or mitigating this bias, it is not always clear whether interventions produce genuinely fairer results. To investigate this issue, we examine how large language models (LLMs) handle gender assignment in a narrative context that is popular, highly ambiguous, and also known to closely reproduce human stereotypes: stories about talking animals. We prompt six leading LLMs to complete an E

Contribution: This paper presents a novel two-phase algorithmic approach that decouples preference satisfaction from fairness optimization in student team formation, achieving both objectives without compromise. The method applies simulated annealing -- a core materials science technique -- to an educational challenge, demonstrating pedagogical integration of administrative processes. Background: Forming effective teams in large engineering cohorts (100+ students) requires balancing student pref

This paper presents an autonomous agentic resource negotiation framework designed to enable zero-touch network slicing in 6G architectures using Large Language Model (LLM) agents. While LLMs offer powerful reasoning capabilities, we demonstrate that such agents inherently suffer from anchoring bias, rigidly adhering to initial heuristic proposals and causing severe network over-provisioning. To systematically mitigate this cognitive bias, we propose a novel randomized anchoring strategy modeled

Data-driven pricing is increasingly prevalent in sectors such as airlines, lending, insurance, and retail. By learning demand models from customer features and setting prices accordingly, these systems may generate discriminatory outcomes that raise fairness concerns. This leads to fundamental questions - how and where should systems incorporate fairness considerations in the pricing pipeline, and how does it ultimately affect societal outcomes? To answer these, we study a stylized model where a

Reinforcement learning from verifiable rewards (RLVR) improves reasoning even when the reward signal is spurious -- assigning credit to the group-plurality answer rather than a ground-truth verifier. Practitioners commonly interpret naive = acc(TRUE) - acc(RANDOM) as the reward-design effect. We prove this estimand is systematically biased: it conflates self-consistency elicitation (sharpening the policy toward its modal answer via majority pseudo-reward) with genuine reward-design signal. Using

Benchmarks are the primary instruments through which AI capability is measured, compared, and governed. This paper argues that the validity of frontier AI benchmarks is a function of the quality of human judgment embedded in their construction, and that this quality is structurally scarce in ways that standard scaling narratives obscure. As foundation models approach ceiling performance on existing evaluation suites, discriminating signal concentrates in the hardest benchmark items, precisely th

When retrieved evidence contradicts parametric memory, language models frequently ignore context and default to memorized priors -- a failure that undermines the core purpose of retrieval augmentation. Contrastive decoding amplifies the context-conditioned output to suppress parametric bias, but existing methods rest on an implicit assumption that this bias is uniform across tokens. A single global contrastive weight over-penalizes safe tokens while leaving genuinely conflicted ones insufficient

Mitigating social bias in Large Language Models (LLMs) presents a distinct alignment challenge: unlike verifiable tasks, bias lacks a single ground truth, creating a high-variance, subjective reward landscape. Previous preference-based fine-tuning methods have major trade-offs: Direct Preference Optimization (DPO) is limited by the lack of exploration inherent in offline training, while Proximal Policy Optimization (PPO) can lead to training instability due to potentially unreliable critic estim

Conventional Post-Training Quantization (PTQ) methods struggle with 4-bit Omni-modal Large Language Models (OLLMs) due to the extreme distribution heterogeneity and disparate outlier patterns across modalities. To address this, we propose MorphoQuant, a modality-aware PTQ framework engineered to preserve cross-modal morphology and mitigate outlier loss. Specifically, we introduce Distribution-Aware Bias Compensation (DABC), which selectively absorbs long-tailed outliers into channel-wise biases.

We present LiveBand, a real-time system that generates high-fidelity music accompaniments to live audio input, respecting strict causal constraints. Our method trains a causal transformer generator in the continuous latent space of a pre-trained causal audio autoencoder, using adversarial sequence-level supervision from a discriminator. At each timestep, the generator receives only the causally available mix context and Gaussian noise, and predicts accompaniment latents without access to future

Machine learning systems deployed in high stakes socioeconomic settings routinely display bias. We formalize bias as a symmetry breaking operation: a classifier is fair if its outputs remain invariant under the counterfactual operation of switching a sensitive attribute, with merit features held fixed. We implement loss based regularization as a symmetry restoring mechanism and evaluate the framework on four synthetic datasets with varying levels of noise, correlation, and bias. The framework ac