Tootfinder

The role of spatial perception in auditory looming bias: neurobehavioral evidence from impossible ears https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2025.1645936/full

Frontiers | The role of spatial perception in auditory looming bias: neurobehavioral evidence from impossible ears
IntroductionSpatial hearing enables both voluntary localization of sound sources and automatic monitoring of the surroundings. The auditory looming bias (ALB...

Sex-based Bias Inherent in the Dice Similarity Coefficient: A Model Independent Analysis for Multiple Anatomical Structures
Hartmut H\"antze, Myrthe Buser, Alessa Hering, Lisa C. Adams, Keno K. Bressem
https://arxiv.org/abs/2509.19778

Sex-based Bias Inherent in the Dice Similarity Coefficient: A Model Independent Analysis for Multiple Anatomical Structures
Overlap-based metrics such as the Dice Similarity Coefficient (DSC) penalize segmentation errors more heavily in smaller structures. As organ size differs by sex, this implies that a segmentation error of equal magnitude may result in lower DSCs in women due to their smaller average organ volumes compared to men. While previous work has examined sex-based differences in models or datasets, no study has yet investigated the potential bias introduced by the DSC itself. This study quantifies sex-b…

Probing Gender Bias in Multilingual LLMs: A Case Study of Stereotypes in Persian
Ghazal Kalhor, Behnam Bahrak
https://arxiv.org/abs/2509.20168 https://arxi…

Probing Gender Bias in Multilingual LLMs: A Case Study of Stereotypes in Persian
Multilingual Large Language Models (LLMs) are increasingly used worldwide, making it essential to ensure they are free from gender bias to prevent representational harm. While prior studies have examined such biases in high-resource languages, low-resource languages remain understudied. In this paper, we propose a template-based probing methodology, validated against real-world data, to uncover gender stereotypes in LLMs. As part of this framework, we introduce the Domain-Specific Gender Skew I…

Non-equilibrium Dynamics of Two-level Systems directly after Cryogenic Alternating Bias
V. Iaia, E. S. Joseph, S. Im, N. Hagopian, S. O'Kelley, C. Kim, N. Materise, S. Patra, V. Lordi, M. A. Eriksson, P. M. Voyles, K. G. Ray, Y. J. Rosen
https://arxiv.org/abs/2509.19223

A bias test for heteroscedastic linear least-squares regression
Eric Blankmeyer
https://arxiv.org/abs/2508.15969 https://arxiv.org/pdf/2508.15969

A bias test for heteroscedastic linear least-squares regression
Linear least squares regression is subject to bias due to an omitted variable, a mismeasured regressor, or simultaneity. A simple test to detect the bias is proposed and explored in simulation and in real data sets.

PediatricsMQA: a Multi-modal Pediatrics Question Answering Benchmark
Adil Bahaj, Mounir Ghogho
https://arxiv.org/abs/2508.16439 https://arxiv.org/pdf/2508.…

PediatricsMQA: a Multi-modal Pediatrics Question Answering Benchmark
Large language models (LLMs) and vision-augmented LLMs (VLMs) have significantly advanced medical informatics, diagnostics, and decision support. However, these models exhibit systematic biases, particularly age bias, compromising their reliability and equity. This is evident in their poorer performance on pediatric-focused text and visual question-answering tasks. This bias reflects a broader imbalance in medical research, where pediatric studies receive less funding and representation despite…

KI-Update kompakt: LLMs für Malware, Anti-Human-Bias, Sutton, Chatbots
Das "KI-Update" liefert werktäglich eine Zusammenfassung der wichtigsten KI-Entwicklungen.
https://www.

FERA: Foil Fencing Referee Assistant Using Pose-Based Multi-Label Move Recognition and Rule Reasoning
Ziwen Chen, Zhong Wang
https://arxiv.org/abs/2509.18527 https://

FERA: Foil Fencing Referee Assistant Using Pose-Based Multi-Label Move Recognition and Rule Reasoning
The sport of fencing, like many other sports, faces challenges in refereeing: subjective calls, human errors, bias, and limited availability in practice environments. We present FERA (Fencing Referee Assistant), a prototype AI referee for foil fencing which integrates pose-based multi-label action recognition and rule-based reasoning. FERA extracts 2D joint positions from video, normalizes them, computes a 101-dimensional kinematic feature set, and applies a Transformer for multi-label move and…

Applications and Challenges of Fairness APIs in Machine Learning Software
Ajoy Das, Gias Uddin, Shaiful Chowdhury, Mostafijur Rahman Akhond, Hadi Hemmati
https://arxiv.org/abs/2508.16377

Applications and Challenges of Fairness APIs in Machine Learning Software
Machine Learning software systems are frequently used in our day-to-day lives. Some of these systems are used in various sensitive environments to make life-changing decisions. Therefore, it is crucial to ensure that these AI/ML systems do not make any discriminatory decisions for any specific groups or populations. In that vein, different bias detection and mitigation open-source software libraries (aka API libraries) are being developed and used. In this paper, we conduct a qualitative study …

Text Takes Over: A Study of Modality Bias in Multimodal Intent Detection
Ankan Mullick, Saransh Sharma, Abhik Jana, Pawan Goyal
https://arxiv.org/abs/2508.16122 https://

Text Takes Over: A Study of Modality Bias in Multimodal Intent Detection
The rise of multimodal data, integrating text, audio, and visuals, has created new opportunities for studying multimodal tasks such as intent detection. This work investigates the effectiveness of Large Language Models (LLMs) and non-LLMs, including text-only and multi-modal models, in the multimodal intent detection task. Our study reveals that Mistral-7B, a text-only LLM, outperforms most competitive multimodal models by approximately 9% on MIntRec-1 and 4% on MIntRec2.0 datasets. This perfor…

AetherCode: Evaluating LLMs' Ability to Win In Premier Programming Competitions
Zihan Wang, Jiaze Chen, Zhicheng Liu, Markus Mak, Yidi Du, Geonsik Moon, Luoqi Xu, Aaron Tua, Kunshuo Peng, Jiayi Lu, Mingfei Xia, Boqian Zou, Chenyang Ran, Guang Tian, Shoutai Zhu, Yeheng Duan, Zhenghui Kang, Zhenxing Lin, Shangshu Li, Qiang Luo, Qingshen Long, Zhiyong Chen, Yihan Xiao, Yurong Wu, Daoguang Zan, Yuyi Fu, Mingxuan Wang, Ming Ding

AetherCode: Evaluating LLMs' Ability to Win In Premier Programming Competitions
Competitive programming has emerged as a critical benchmark for evaluating the reasoning and coding capabilities of Large Language Models (LLMs). Despite impressive progress on existing benchmarks, we argue that current evaluations overstate model proficiency, masking a substantial gap between LLMs and elite human programmers. This gap arises from two key limitations: insufficient difficulty and scope of benchmark problems, and evaluation bias from low-quality test cases. To address these short…

DoubleGen: Debiased Generative Modeling of Counterfactuals
Alex Luedtke, Kenji Fukumizu
https://arxiv.org/abs/2509.16842 https://arxiv.org/pdf/2509.16842…

DoubleGen: Debiased Generative Modeling of Counterfactuals
Generative models for counterfactual outcomes face two key sources of bias. Confounding bias arises when approaches fail to account for systematic differences between those who receive the intervention and those who do not. Misspecification bias arises when methods attempt to address confounding through estimation of an auxiliary model, but specify it incorrectly. We introduce DoubleGen, a doubly robust framework that modifies generative modeling training objectives to mitigate these biases. Th…

Bias in the Picture: Benchmarking VLMs with Social-Cue News Images and LLM-as-Judge Assessment
Aravind Narayanan, Vahid Reza Khazaie, Shaina Raza
https://arxiv.org/abs/2509.19659

Bias in the Picture: Benchmarking VLMs with Social-Cue News Images and LLM-as-Judge Assessment
Large vision-language models (VLMs) can jointly interpret images and text, but they are also prone to absorbing and reproducing harmful social stereotypes when visual cues such as age, gender, race, clothing, or occupation are present. To investigate these risks, we introduce a news-image benchmark consisting of 1,343 image-question pairs drawn from diverse outlets, which we annotated with ground-truth answers and demographic attributes (age, gender, race, occupation, and sports). We evaluate a…

Adaptive User Interest Modeling via Conditioned Denoising Diffusion For Click-Through Rate Prediction
Qihang Zhao, Xiaoyang Zheng, Ben Chen, Zhongbo Sun, Chenyi Lei
https://arxiv.org/abs/2509.19876

Adaptive User Interest Modeling via Conditioned Denoising Diffusion For Click-Through Rate Prediction
User behavior sequences in search systems resemble "interest fossils", capturing genuine intent yet eroded by exposure bias, category drift, and contextual noise. Current methods predominantly follow an "identify-aggregate" paradigm, assuming sequences immutably reflect user preferences while overlooking the organic entanglement of noise and genuine interest. Moreover, they output static, context-agnostic representations, failing to adapt to dynamic intent shifts under varying Query-User-Item-C…

Ensuring Reliable Participation in Subjective Video Quality Tests Across Platforms
Babak Naderi, Ross Cutler
https://arxiv.org/abs/2509.20001 https://arxiv…

Ensuring Reliable Participation in Subjective Video Quality Tests Across Platforms
Subjective video quality assessment (VQA) is the gold standard for measuring end-user experience across communication, streaming, and UGC pipelines. Beyond high-validity lab studies, crowdsourcing offers accurate, reliable, faster, and cheaper evaluation-but suffers from unreliable submissions by workers who ignore instructions or game rewards. Recent tests reveal sophisticated exploits of video metadata and rising use of remote-desktop (RD) connections, both of which bias results. We propose o…

#SteadySupporter
Wie neutral sind KI-Systeme bei der medizinischen Diagnostik?
Eine neue Studie zeigt, dass selbst auf Expertenniveau arbeitende Systeme #Krankheiten übersehen und dabei marginalisierte Gruppen benachteiligen können.

Bias in der Medizin: Wenn KI-Röntgen-Diagnostik nicht neutral ist
Wie fair ist KI in der Medizin wirklich? Diese Studie offenbart Verzerrungen in der Diagnose marginalisierter Gruppen.

Tree-based methods for length-biased survival data
Jinwoo Lee, Jiyu Sun, Hyunwoo Lee, Donghwan Lee
https://arxiv.org/abs/2508.16312 https://arxiv.org/pdf/2…

Tree-based methods for length-biased survival data
Left-truncated survival data commonly arise in prevalent cohort studies, where only individuals who have experienced disease onset and survived until enrollment in the study. When the onset process follows a stationary Poisson process, the resulting data are length-biased. This sampling mechanism induces a selection bias towards longer survival individuals, and nonparametric and semiparametric methods for traditional survival data are not directly applicable. While tree-based methods developed …

The Pareto Frontier of Resilient Jet Tagging
Rikab Gambhir, Matt LeBlanc, Yuanchen Zhou
https://arxiv.org/abs/2509.19431 https://arxiv.org/pdf/2509.19431…

The Pareto Frontier of Resilient Jet Tagging
Classifying hadronic jets using their constituents' kinematic information is a critical task in modern high-energy collider physics. Often, classifiers are designed by targeting the best performance using metrics such as accuracy, AUC, or rejection rates. However, the use of a single metric can lead to the use of architectures that are more model-dependent than competitive alternatives, leading to potential uncertainty and bias in analysis. We explore such trade-offs and demonstrate the consequ…

As beautiful as a sailboat looks. In my eyes it's totally not worth the effort to own one (on lakes in Germany). While I enjoy helping my friends with their boats occasionally ... the amount of maintenance just confirms my bias every time :D

Geometrical portrait of Multipath error propagation in GNSS Direct Position Estimation
Jihong Huang, Rong Yang, Wei Gao, Xingqun Zhan, Zheng Yao
https://arxiv.org/abs/2507.18096

Geometrical portrait of Multipath error propagation in GNSS Direct Position Estimation
Direct Position Estimation (DPE) is a method that directly estimate position, velocity, and time (PVT) information from cross ambiguity function (CAF) of the GNSS signals, significantly enhancing receiver robustness in urban environments. However, there is still a lack of theoretical characterization on multipath errors in the context of DPE theory. Geometric observations highlight the unique characteristics of DPE errors stemming from multipath and thermal noise as estimation bias and variance…

Simulation-Based Inference for Direction Reconstruction of Ultra-High-Energy Cosmic Rays with Radio Arrays
Oscar Macias, Zachary Mason, Matthew Ho, Ars\`ene Ferri\`ere, Aur\'elien Benoit-L\'evy, Mat\'ias Tueros
https://arxiv.org/abs/2508.15991

Simulation-Based Inference for Direction Reconstruction of Ultra-High-Energy Cosmic Rays with Radio Arrays
Ultra-high-energy cosmic-ray (UHECR) observatories require unbiased direction reconstruction to enable multi-messenger astronomy with sparse, nanosecond-scale radio pulses. Explicit likelihood methods often rely on simplified models, which may bias results and understate uncertainties. We introduce a simulation-based inference pipeline that couples a physics-informed graph neural network (GNN) to a normalizing-flow posterior within the \textit{Learning the Universe Implicit Likelihood Inference…

Testing the Constancy of Type Ia Supernova Luminosities with Gaussian Process
Akshay Rana
https://arxiv.org/abs/2509.19494 https://arxiv.org/pdf/2509.19494…

Testing the Constancy of Type Ia Supernova Luminosities with Gaussian Process
Type Ia supernovae (SNe~Ia) are central to studies of cosmic expansion, under the assumption that their absolute magnitude $M_B$ does not evolve with redshift. Even small drifts in brightness can bias cosmological parameters such as $H_0$ and $w$. Here we test this assumption using a non-parametric Gaussian Process (GP) reconstruction of the expansion history from cosmic chronometer $H(z)$ data, which provides a model-independent baseline distance modulus, $μ_{\rm GP}(z)$. To propagate uncerta…

2025 Southeast Asia Eleven Nations Influence Index Report
Wei Meng
https://arxiv.org/abs/2509.19953 https://arxiv.org/pdf/2509.19953

2025 Southeast Asia Eleven Nations Influence Index Report
This study constructs a fully data-driven and reproducible Southeast Asia Influence Index (SAII v3) to reduce bias from expert scoring and subjective weighting while mapping hierarchical power structures across the eleven ASEAN nations. We aggregate authoritative open-source indicators across four dimensions (economic, military, diplomatic, socio-technological) and apply a three-tiered standardization chain quantile-Box-Cox-min-max to mitigate outliers and skewness. Weights are obtained through…

AJAHR: Amputated Joint Aware 3D Human Mesh Recovery
Hyunjin Cho, Giyun Choi, Jongwon Choi
https://arxiv.org/abs/2509.19939 https://arxiv.org/pdf/2509.19939…

AJAHR: Amputated Joint Aware 3D Human Mesh Recovery
Existing human mesh recovery methods assume a standard human body structure, overlooking diverse anatomical conditions such as limb loss. This assumption introduces bias when applied to individuals with amputations - a limitation further exacerbated by the scarcity of suitable datasets. To address this gap, we propose Amputated Joint Aware 3D Human Mesh Recovery (AJAHR), which is an adaptive pose estimation framework that improves mesh reconstruction for individuals with limb loss. Our model in…

Shanks bias in function fields
Seewoo Lee
https://arxiv.org/abs/2509.16142 https://arxiv.org/pdf/2509.16142

Shanks bias in function fields
We study the function field analogue of Shanks bias. For Liouville function $λ(f)$, we compare the number of monic polynomials $f$ with $λ(f) χ_m(f) = 1$ and $λ(f) χ_m(f) = -1$ for a nontrivial quadratic character $χ_m$ modulo a monic square-free polynomial $m$ over a finite field. Under Grand Simplicity Hypothesis (GSH) for $L$-functions, we prove that $λ\cdot χ_m$ is biased towards $+1$. We also give some examples where GSH is violated.

Evaluation-Aware Reinforcement Learning
Shripad Vilasrao Deshmukh, Will Schwarzer, Scott Niekum
https://arxiv.org/abs/2509.19464 https://arxiv.org/pdf/2509…

Evaluation-Aware Reinforcement Learning
Policy evaluation is often a prerequisite for deploying safety- and performance-critical systems. Existing evaluation approaches frequently suffer from high variance due to limited data and long-horizon tasks, or high bias due to unequal support or inaccurate environmental models. We posit that these challenges arise, in part, from the standard reinforcement learning (RL) paradigm of policy learning without explicit consideration of evaluation. As an alternative, we propose evaluation-aware rei…

Uncertainty Quantification for Evaluating Machine Translation Bias
Ieva Raminta Stali\=unait\.e, Julius Cheng, Andreas Vlachos
https://arxiv.org/abs/2507.18338 https://

Uncertainty Quantification for Evaluating Machine Translation Bias
In machine translation (MT), when the source sentence includes a lexeme whose gender is not overtly marked, but whose target-language equivalent requires gender specification, the model must infer the appropriate gender from the context and/or external knowledge. Studies have shown that MT models exhibit biased behaviour, relying on stereotypes even when they clash with contextual information. We posit that apart from confidently translating using the correct gender when it is evident from the …

Survivors, Complainers, and Borderliners: Upward Bias in Online Discussions of Academic Conference Reviews
Hangxiao Zhu, Yian Yin, Yu Zhang
https://arxiv.org/abs/2509.16831 http…

Survivors, Complainers, and Borderliners: Upward Bias in Online Discussions of Academic Conference Reviews
Online discussion platforms, such as community Q&A sites and forums, have become important hubs where academic conference authors share and seek information about the peer review process and outcomes. However, these discussions involve only a subset of all submissions, raising concerns about the representativeness of the self-reported review scores. In this paper, we conduct a systematic study comparing the review score distributions of self-reported submissions in online discussions (based on …

Multichannel highly sensitive diamond quantum magnetometer
Atsumi Yoshimura, Ayumi Kanamoto, Naota Sekiguchi, Chikara Shinei, Masashi Miyakawa, Takashi Taniguchi, Tokuyuki Teraji, Hiroshi Abe, Shinobu Onoda, Takeshi Ohshima, Takayuki Iwasaki, Mutsuko Hatano
https://arxiv.org/abs/2509.20055

Multichannel highly sensitive diamond quantum magnetometer
We demonstrate a highly sensitive real-time magnetometry method at two measurement points. This magnetometry method is based on the frequency-division multiplexing of continuous-wave optically detected magnetic resonance. We use two ensembles of nitrogen-vacancy (NV) centers separated by 3.6 mm to measure a magnetic field. A different bias field is applied to the two NV ensembles to resolve the resonance peak for each ensemble in the frequency space and enables the multiplexed magnetometry at t…

Evaluating Bias Reduction Methods in Binary Emax Model for Reliable Dose-Response Estimation
Jiangshan Zhang, Vivek Pradhan, Yuxi Zhao
https://arxiv.org/abs/2509.18459 https://

KI-Systeme bevorzugen eigene Texte: Studie warnt vor "Anti-Human-Bias"
Wissenschaftler haben nachgewiesen, dass Large Language Models Inhalte anderer KI-Systeme bevorzugen. Was Menschen daraus lernen können.

I like to play around as an anonymous commenter in online newspaper columns. If you point out the biases of #AI systems, the comment gets deleted because it is considered too polemical.
The comment was addressing an article about AI in public service and the use in refugee applications 🫣

Thirteen newsroom leaders and ethicists discuss elements of journalistic ethics, including dealing with AI, anonymous leaks, bias, lies, and audience needs (Columbia Journalism Review)
https://www.cjr.org/feature/thirteen-journ

Thirteen Journalists on How They Are Rethinking Ethics 
We asked newsroom leaders and ethicists what they’re keeping or changing in an era of Trump, “fake news,” AI, and industry decline.

Remote coding interviews where you have to use your own machine have so much implicit bias built into them. As I haven't had a personal OS X machine in a decade, I have to use my Linux box which leads to...interesting shenanigans.

Enhanced Survival Trees
Ruiwen Zhou, Ke Xie, Lei Liu, Zhichen Xu, Jimin Ding, Xiaogang Su
https://arxiv.org/abs/2509.18494 https://arxiv.org/pdf/2509.18494…

Enhanced Survival Trees
We introduce a new survival tree method for censored failure time data that incorporates three key advancements over traditional approaches. First, we develop a more computationally efficient splitting procedure that effectively mitigates the end-cut preference problem, and we propose an intersected validation strategy to reduce the variable selection bias inherent in greedy searches. Second, we present a novel framework for determining tree structures through fused regularization. In combinati…

Bias-variance Tradeoff in Tensor Estimation
Shivam Kumar, Haotian Xu, Carlos Misael Madrid Padilla, Yuehaw Khoo, Oscar Hernan Madrid Padilla, Daren Wang
https://arxiv.org/abs/2509.17382

Bias-variance Tradeoff in Tensor Estimation
We study denoising of a third-order tensor when the ground-truth tensor is not necessarily Tucker low-rank. Specifically, we observe $$ Y=X^\ast+Z\in \mathbb{R}^{p_{1} \times p_{2} \times p_{3}}, $$ where $X^\ast$ is the ground-truth tensor, and $Z$ is the noise tensor. We propose a simple variant of the higher-order tensor SVD estimator $\widetilde{X}$. We show that uniformly over all user-specified Tucker ranks $(r_{1},r_{2},r_{3})$, $$ \| \widetilde{X} - X^* \|_{ \mathrm{F}}^2 = O …

BiTAA: A Bi-Task Adversarial Attack for Object Detection and Depth Estimation via 3D Gaussian Splatting
Yixun Zhang, Feng Zhou, Jianqin Yin
https://arxiv.org/abs/2509.19793 http…

BiTAA: A Bi-Task Adversarial Attack for Object Detection and Depth Estimation via 3D Gaussian Splatting
Camera-based perception is critical to autonomous driving yet remains vulnerable to task-specific adversarial manipulations in object detection and monocular depth estimation. Most existing 2D/3D attacks are developed in task silos, lack mechanisms to induce controllable depth bias, and offer no standardized protocol to quantify cross-task transfer, leaving the interaction between detection and depth underexplored. We present BiTAA, a bi-task adversarial attack built on 3D Gaussian Splatting th…

Bridging the gap between training and inference in LM-based TTS models
Ruonan Zhang, Lingzhou Mu, Xixin Wu, Kai Zhang
https://arxiv.org/abs/2509.17021 https://

Bridging the gap between training and inference in LM-based TTS models
Recent advancements in text-to-speech (TTS) have shown that language model (LM) based systems offer competitive performance compared to traditional approaches. However, in training, TTS models use ground-truth (GT) tokens as prefixes to predict the next token, while in inference these tokens are not available, a gap between training and inference that is often neglected. In this study, we propose a prompt-guided hybrid training scheme to mitigate exposure bias in popular LM-based TTS systems. O…

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[1/3]:
- Causally Testing Gender Bias in LLMs: A Case Study on Occupational Bias
Yuen Chen, Vethavikashini Chithrra Raghuram, Justus Mattern, Rada Mihalcea, Zhijing Jin

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[5/12]:
- Auto-Search and Refinement: An Automated Framework for Gender Bias Mitigation in Large Language M...
Yue Xu, Chengyan Fu, Li Xiong, Sibei Yang, Wenjie Wang

Reading Between the Lines: A Study of Thematic Bias in Book Recommender Systems
Nityaa Kalra, Savvina Daniil
https://arxiv.org/abs/2508.15643 https://arxiv…

Reading Between the Lines: A Study of Thematic Bias in Book Recommender Systems
Recommender systems help users discover new content, but can also reinforce existing biases, leading to unfair exposure and reduced diversity. This paper introduces and investigates thematic bias in book recommendations, defined as a disproportionate favouring or neglect of certain book themes. We adopt a multi-stage bias evaluation framework using the Book-Crossing dataset to evaluate thematic bias in recommendations and its impact on different user groups. Our findings show that thematic bi…

EvoFormer: Learning Dynamic Graph-Level Representations with Structural and Temporal Bias Correction
Haodi Zhong, Liuxin Zou, Di Wang, Bo Wang, Zhenxing Niu, Quan Wang
https://arxiv.org/abs/2508.15378 …

EvoFormer: Learning Dynamic Graph-Level Representations with Structural and Temporal Bias Correction
Dynamic graph-level embedding aims to capture structural evolution in networks, which is essential for modeling real-world scenarios. However, existing methods face two critical yet under-explored issues: Structural Visit Bias, where random walk sampling disproportionately emphasizes high-degree nodes, leading to redundant and noisy structural representations; and Abrupt Evolution Blindness, the failure to effectively detect sudden structural changes due to rigid or overly simplistic temporal m…

Crosslisted article(s) found for cs.DL. https://arxiv.org/list/cs.DL/new
[1/1]:
- Survivors, Complainers, and Borderliners: Upward Bias in Online Discussions of Academic Conferenc...
Hangxiao Zhu, Yian Yin, Yu Zhang

Designing Psychometric Bias Measures for ChatBots: An Application to Racial Bias Measurement
Mouhacine Benosman
https://arxiv.org/abs/2509.13324 https://ar…

Designing Psychometric Bias Measures for ChatBots: An Application to Racial Bias Measurement
Artificial intelligence (AI), particularly in the form of large language models (LLMs) or chatbots, has become increasingly integrated into our daily lives. In the past five years, several LLMs have been introduced, including ChatGPT by OpenAI, Claude by Anthropic, and Llama by Meta, among others. These models have the potential to be employed across a wide range of human-machine interaction applications, such as chatbots for information retrieval, assistance in corporate hiring decisions, coll…

Fast and accurate Gaia-unWISE quasar mock catalogs from LPT and Eulerian bias
Francesco Sinigaglia, Francisco-Shu Kitaura, Mahlet Shiferaw, Ginevra Favole, Kate Storey-Fisher, Nestor Arsenov
https://arxiv.org/abs/2509.15890

Fast and accurate Gaia-unWISE quasar mock catalogs from LPT and Eulerian bias
We present $100$ full-sky quasar spectrophotometric mock catalogs with smooth redshift evolution from $z=0$ to $z\sim 4$, tailored to analyze the Gaia-unWISE Quasar Catalog (Quaia). In particular, we apply a novel hierarchical nonlocal nonlinear bias scheme (Hicobian) to dark matter fields generated through Augmented Lagrangian Perturbation Theory on the lightcone (WebON code), calibrating the free parameters of the bias model on Abacus quasar HOD mock catalogs tuned to reproduce DESI Early Dat…

The Narcissus Hypothesis:Descending to the Rung of Illusion
Riccardo Cadei, Christian Intern\`o
https://arxiv.org/abs/2509.17999 https://arxiv.org/pdf/2509…

The Narcissus Hypothesis:Descending to the Rung of Illusion
Modern foundational models increasingly reflect not just world knowledge, but patterns of human preference embedded in their training data. We hypothesize that recursive alignment-via human feedback and model-generated corpora-induces a social desirability bias, nudging models to favor agreeable or flattering responses over objective reasoning. We refer to it as the Narcissus Hypothesis and test it across 31 models using standardized personality assessments and a novel Social Desirability Bias …

“Developers on teams with high AI adoption complete 21% more tasks and merge 98% more pull requests, but PR review time increases 91%, revealing a critical bottleneck: human approval.“
Maybe it’s confirmation bias, but I can see that. You generate more, maybe harder to comprehend, code that still has to be double checked by people who weren’t involved in the process. That slows you down unless you ignore understanding by, you guessed it, moving fast and breaking things.

The AI Productivity Paradox Research Report | Faros AI
Key findings from the AI Productivity Paradox Report 2025. Research reveals AI coding assistants increase developer output, but not company productivity. Uncover strategies and enablers for a measurable return on investment.

Es wurde ja erwartet, aber tatsächlich fand ich den Punkt der Verstärkung bestehender Ungleichheiten bislang stark unterrepräsentiert in der Berichterstattung.
KI-Bias bei Gehaltsratschlägen: Wie Sprachmodelle Frauen systematisch benachteiligen - Notebookcheck.com News

C$^2$MIL: Synchronizing Semantic and Topological Causalities in Multiple Instance Learning for Robust and Interpretable Survival Analysis
Min Cen, Zhenfeng Zhuang, Yuzhe Zhang, Min Zeng, Baptiste Magnier, Lequan Yu, Hong Zhang, Liansheng Wang
https://arxiv.org/abs/2509.20152

C$^2$MIL: Synchronizing Semantic and Topological Causalities in Multiple Instance Learning for Robust and Interpretable Survival Analysis
Graph-based Multiple Instance Learning (MIL) is widely used in survival analysis with Hematoxylin and Eosin (H\&E)-stained whole slide images (WSIs) due to its ability to capture topological information. However, variations in staining and scanning can introduce semantic bias, while topological subgraphs that are not relevant to the causal relationships can create noise, resulting in biased slide-level representations. These issues can hinder both the interpretability and generalization of the …

An investigation finds rampant caste bias in ChatGPT and Sora; a researcher also finds caste bias in Sarvam AI, which touts itself as a sovereign AI for India (Nilesh Christopher/MIT Technology Review)
https://www.technologyreview.com/2025/10/01/1124621/openai-in…

OpenAI is huge in India. Its models are steeped in caste bias.
India is OpenAI’s second-largest market, but ChatGPT and Sora reproduce caste stereotypes that harm millions of people.

$\Delta_T$ Noise as a Robust Diagnostic for Chiral, Helical and Trivial Edge Modes
Sachiraj Mishra, Colin Benjamin
https://arxiv.org/abs/2509.16747 https://

$Δ_T$ Noise as a Robust Diagnostic for Chiral, Helical and Trivial Edge Modes
In this article we demonstrate that $Δ_T$ noise provides a sensitive, practical probe for distinguishing chiral edge modes from topological helical and trivial (non-topological) helical edge transport. Measured under zero-current conditions, $Δ_T$ noise reveals contrasts that conventional conductance measurements typically miss. Crucially, $Δ_T$ requires no external energy input in the form of an applied voltage bias, yet encodes the same intrinsic information that shot noise yields in the z…

Position Bias Mitigates Position Bias:Mitigate Position Bias Through Inter-Position Knowledge Distillation
Yifei Wang, Feng Xiong, Yong Wang, Linjing Li, Xiangxiang Chu, Daniel Dajun Zeng
https://arxiv.org/abs/2508.15709

Position Bias Mitigates Position Bias:Mitigate Position Bias Through Inter-Position Knowledge Distillation
Positional bias (PB), manifesting as non-uniform sensitivity across different contextual locations, significantly impairs long-context comprehension and processing capabilities. While prior work seeks to mitigate PB through modifying the architectures causing its emergence, significant PB still persists. To address PB effectively, we introduce \textbf{Pos2Distill}, a position to position knowledge distillation framework. Pos2Distill transfers the superior capabilities from advantageous position…

Surprise (non)
"Les outils d'intelligence artificielle utilisés par les médecins risquent d'entraîner une dégradation de l'état de santé des femmes et des minorités ethniques.
En effet, de plus en plus d'études montrent que de nombreux modèles de langage simplifiés minimisent les symptômes de ces patients."
#IA

Mitigating Strategy-Selection Bias in Reasoning for More Effective Test-Time Scaling
Zongqian Wu, Baoduo Xu, Tianyu Li, Zhu Sun, Xiaofeng Zhu, Lei Feng
https://arxiv.org/abs/2509.17905

Mitigating Strategy-Selection Bias in Reasoning for More Effective Test-Time Scaling
Test-time scaling (TTS) has been shown to improve the performance of large language models (LLMs) by sampling and aggregating diverse reasoning paths. However, existing research has overlooked a critical issue: selection bias of reasoning strategies during scaling. Specifically, when generating reasoning processes, LLMs tend to follow certain strategies (e.g., algebraic solutions for math problems) while neglecting other valid alternatives (e.g., geometric solutions), resulting in insufficient …

Evidence Backs Trump on Higher Ed's Bias (Wall Street Journal)
https://www.wsj.com/opinion/evidence-backs-trump-on-higher-eds-bias-politics-13d4fec0
http://www.memeorandum.com/250813/p76#a250813p76

Multi-state Models For Modeling Disease Histories Based On Longitudinal Data
Simon Wiegrebe, Johannes Piller, Mathias Gorski, Merle Behr, Helmut K\"uchenhoff, Iris M. Heid, Andreas Bender
https://arxiv.org/abs/2509.19956

Multi-state Models For Modeling Disease Histories Based On Longitudinal Data
Multi-stage disease histories derived from longitudinal data are becoming increasingly available as registry data and biobanks expand. Multi-state models are suitable to investigate transitions between different disease stages in presence of competing risks. In this context, however their estimation is complicated by dependent left-truncation, multiple time scales, index event bias, and interval-censoring. In this work, we investigate the extension of piecewise exponential additive models (PAMs…

LoCaL: Countering Surface Bias in Code Evaluation Metrics
Simantika Bhattacharjee Dristi, Matthew B. Dwyer
https://arxiv.org/abs/2509.15397 https://arxiv.o…

LoCaL: Countering Surface Bias in Code Evaluation Metrics
With the increasing popularity of large language models (LLMs) and LLM-based agents, reliable and effective code evaluation metrics (CEMs) have become crucial for progress across several software engineering tasks. While popular benchmarks often provide test cases to assess the correctness of generated code, crafting and executing test cases is expensive. Reference-based CEMs provide a cheaper alternative by scoring a candidate program based on its functional similarity to a reference. Although…

MoTiC: Momentum Tightness and Contrast for Few-Shot Class-Incremental Learning
Zeyu He, Shuai Huang, Yuwu Lu, Ming Zhao
https://arxiv.org/abs/2509.19664 https://

MoTiC: Momentum Tightness and Contrast for Few-Shot Class-Incremental Learning
Few-Shot Class-Incremental Learning (FSCIL) must contend with the dual challenge of learning new classes from scarce samples while preserving old class knowledge. Existing methods use the frozen feature extractor and class-averaged prototypes to mitigate against catastrophic forgetting and overfitting. However, new-class prototypes suffer significant estimation bias due to extreme data scarcity, whereas base-class prototypes benefit from sufficient data. In this work, we theoretically demonstra…

Power Spectral Density Estimation via Universal Truncated Order Statistics Filtering
David Campos Anchieta, John R. Buck
https://arxiv.org/abs/2509.16359 https://

Power Spectral Density Estimation via Universal Truncated Order Statistics Filtering
Loud transient signals in underwater acoustic data increase the bias and variance of background noise power spectral density (PSD) estimates based on sample mean. Recently, two PSD estimators mitigated the loud transient impact on PSD estimates by applying order statistics filtering (OSF). The first, the Schwock and Abadi Welch Percentile, scales a single rank order statistic (OS) of consecutive periodograms. The second, the truncated linear order statistics filter, is a weighted sum of OS up t…

BEFT: Bias-Efficient Fine-Tuning of Language Models
Baichuan Huang, Ananth Balashankar, Amir Aminifar
https://arxiv.org/abs/2509.15974 https://arxiv.org/pd…

BEFT: Bias-Efficient Fine-Tuning of Language Models
Fine-tuning all-bias-terms stands out among various parameter-efficient fine-tuning (PEFT) techniques, owing to its out-of-the-box usability and competitive performance, especially in low-data regimes. Bias-only fine-tuning has the potential for unprecedented parameter efficiency. However, the link between fine-tuning different bias terms (i.e., bias terms in the query, key, or value projections) and downstream performance remains unclear. The existing approaches, e.g., based on the magnitude o…

Augmenting Limited and Biased RCTs through Pseudo-Sample Matching-Based Observational Data Fusion Method
Kairong Han, Weidong Huang, Taiyang Zhou, Peng Zhen, Kun Kuang
https://arxiv.org/abs/2509.18148 …

Augmenting Limited and Biased RCTs through Pseudo-Sample Matching-Based Observational Data Fusion Method
In the online ride-hailing pricing context, companies often conduct randomized controlled trials (RCTs) and utilize uplift models to assess the effect of discounts on customer orders, which substantially influences competitive market outcomes. However, due to the high cost of RCTs, the proportion of trial data relative to observational data is small, which only accounts for 0.65\% of total traffic in our context, resulting in significant bias when generalizing to the broader user base. Addition…

Bias is a Math Problem, AI Bias is a Technical Problem: 10-year Literature Review of AI/LLM Bias Research Reveals Narrow [Gender-Centric] Conceptions of 'Bias', and Academia-Industry Gap
Sourojit Ghosh, Kyra Wilson
https://arxiv.org/abs/2508.11067

Bias is a Math Problem, AI Bias is a Technical Problem: 10-year Literature Review of AI/LLM Bias Research Reveals Narrow [Gender-Centric] Conceptions of 'Bias', and Academia-Industry Gap
The rapid development of AI tools and implementation of LLMs within downstream tasks has been paralleled by a surge in research exploring how the outputs of such AI/LLM systems embed biases, a research topic which was already being extensively explored before the era of ChatGPT. Given the high volume of research around the biases within the outputs of AI systems and LLMs, it is imperative to conduct systematic literature reviews to document throughlines within such research. In this paper, we c…

Improving Fairness in Graph Neural Networks via Counterfactual Debiasing
Zengyi Wo, Chang Liu, Yumeng Wang, Minglai Shao, Wenjun Wang
https://arxiv.org/abs/2508.14683 https://…

Improving Fairness in Graph Neural Networks via Counterfactual Debiasing
Graph Neural Networks (GNNs) have been successful in modeling graph-structured data. However, similar to other machine learning models, GNNs can exhibit bias in predictions based on attributes like race and gender. Moreover, bias in GNNs can be exacerbated by the graph structure and message-passing mechanisms. Recent cutting-edge methods propose mitigating bias by filtering out sensitive information from input or representations, like edge dropping or feature masking. Yet, we argue that such st…

State of Abdominal CT Datasets: A Critical Review of Bias, Clinical Relevance, and Real-world Applicability
Saeide Danaei, Zahra Dehghanian, Elahe Meftah, Nariman Naderi, Seyed Amir Ahmad Safavi-Naini, Faeze Khorasanizade, Hamid R. Rabiee
https://arxiv.org/abs/2508.13626

State of Abdominal CT Datasets: A Critical Review of Bias, Clinical Relevance, and Real-world Applicability
This systematic review critically evaluates publicly available abdominal CT datasets and their suitability for artificial intelligence (AI) applications in clinical settings. We examined 46 publicly available abdominal CT datasets (50,256 studies). Across all 46 datasets, we found substantial redundancy (59.1\% case reuse) and a Western/geographic skew (75.3\% from North America and Europe). A bias assessment was performed on the 19 datasets with >=100 cases; within this subset, the most preval…

AnchoredAI: Contextual Anchoring of AI Comments Improves Writer Agency and Ownership
Martin Lou, Jackie Crowley, Samuel Dodson, Dongwook Yoon
https://arxiv.org/abs/2509.16128 ht…

AnchoredAI: Contextual Anchoring of AI Comments Improves Writer Agency and Ownership
Generative AI is increasingly integrated into writing support, yet current chat-based interfaces often obscure referential context and risk amplifying automation bias and overreliance. We introduce AnchoredAI, a novel system that anchors AI feedback directly to relevant text spans. AnchoredAI implements two key mechanisms: (1) an Anchoring Context Window (ACW) that maintains unique, context-rich references, and (2) an update-aware context retrieval method that preserves the intent of prior comm…

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[1/3]:
- Prompting Techniques for Reducing Social Bias in LLMs through System 1 and System 2 Cognitive Pro...
Mahammed Kamruzzaman, Gene Louis Kim

How the outrage over Jimmy Kimmel's remarks on Charlie Kirk ballooned, starting with one X user, who monitors late night shows for liberal bias, posting a clip (Stuart A. Thompson/New York Times)
https://www.nytimes.com/2025/09/19/technology/kimmel-carr-outr…

How Outrage at Kimmel Grew to a Shout From a Whisper
Right-wing users on social media on Tuesday were frustrated but not yet apoplectic about Jimmy Kimmel’s monologue. Things changed.

Ethical Considerations of Large Language Models in Game Playing
Qingquan Zhang, Yuchen Li, Bo Yuan, Julian Togelius, Georgios N. Yannakakis, Jialin Liu
https://arxiv.org/abs/2508.16065

Ethical Considerations of Large Language Models in Game Playing
Large language models (LLMs) have demonstrated tremendous potential in game playing, while little attention has been paid to their ethical implications in those contexts. This work investigates and analyses the ethical considerations of applying LLMs in game playing, using Werewolf, also known as Mafia, as a case study. Gender bias, which affects game fairness and player experience, has been observed from the behaviour of LLMs. Some roles, such as the Guard and Werewolf, are more sensitive than…

Estimation-Theoretic Bias Reduction for Oscillometric Blood Pressure Readings
Masoud Nateghi, Reza Sameni
https://arxiv.org/abs/2508.15687 https://arxiv.or…

Estimation-Theoretic Bias Reduction for Oscillometric Blood Pressure Readings
Oscillometry is the standard method for non-invasive, cuff-based blood pressure (BP) measurement, but it introduces systematic errors that may impact clinical accuracy. This study investigates the sources of these errors--primarily the limitations of oscillometry itself and respiration-induced fluctuations--using BP waveform data from the MIMIC database. Oscillometry tends to underestimate systolic BP and overestimate diastolic BP, while respiration introduces cyclical variations that further d…

In response to Elon Musk's claims that the App Store favors the ChatGPT app, Apple says the App Store "is designed to be fair and free of bias" (Mark Gurman/@markgurman)
https://x.com/markgurman/status/1955383759853007198

Mark Gurman (@markgurman) on X
Apple on @elonmusk’s claims: "The App Store is designed to be fair and free of bias. We feature thousands of apps through charts, algorithmic recommendations, and curated lists selected by experts using objective criteria.” https://t.co/w4NFINOF9x

Trumps anti-woke-Verordnung: Meta arbeitet mit rechtem Influencer
Um den vermeintlich woken Bias aus KI-Anwendungen zu bekommen, arbeitet Meta nun mit dem konservativen Robby Starbuck zusammen.
https://www.…

Assessing Trustworthiness of AI Training Dataset using Subjective Logic -- A Use Case on Bias
Koffi Ismael Ouattara, Ioannis Krontiris, Theo Dimitrakos, Frank Kargl
https://arxiv.org/abs/2508.13813

Assessing Trustworthiness of AI Training Dataset using Subjective Logic -- A Use Case on Bias
As AI systems increasingly rely on training data, assessing dataset trustworthiness has become critical, particularly for properties like fairness or bias that emerge at the dataset level. Prior work has used Subjective Logic to assess trustworthiness of individual data, but not to evaluate trustworthiness properties that emerge only at the level of the dataset as a whole. This paper introduces the first formal framework for assessing the trustworthiness of AI training datasets, enabling uncert…

Confirmation Bias as a Cognitive Resource in LLM-Supported Deliberation
Sander de Jong, Rune M{\o}berg Jacobsen, Niels van Berkel
https://arxiv.org/abs/2509.14824 https://

Confirmation Bias as a Cognitive Resource in LLM-Supported Deliberation
Large language models (LLMs) are increasingly used in group decision-making, but their influence risks fostering conformity and reducing epistemic vigilance. Drawing on the Argumentative Theory of Reasoning, we argue that confirmation bias, often seen as detrimental, can be harnessed as a resource when paired with critical evaluation. We propose a three-step process in which individuals first generate ideas independently, then use LLMs to refine and articulate them, and finally engage with LLMs…

Investigating Bias: A Multilingual Pipeline for Generating, Solving, and Evaluating Math Problems with LLMs
Mariam Mahran, Katharina Simbeck
https://arxiv.org/abs/2509.17701 htt…

Investigating Bias: A Multilingual Pipeline for Generating, Solving, and Evaluating Math Problems with LLMs
Large Language Models (LLMs) are increasingly used for educational support, yet their response quality varies depending on the language of interaction. This paper presents an automated multilingual pipeline for generating, solving, and evaluating math problems aligned with the German K-10 curriculum. We generated 628 math exercises and translated them into English, German, and Arabic. Three commercial LLMs (GPT-4o-mini, Gemini 2.5 Flash, and Qwen-plus) were prompted to produce step-by-step solu…

Where does non-Universality in Assembly Bias come from?
Charuhas Shiveshwarkar, Marilena Loverde, Christopher M. Hirata, Drew Jamieson
https://arxiv.org/abs/2508.11798 https://

Where does non-Universality in Assembly Bias come from?
Constraints on local primordial non-Gaussianity (LPnG) obtained from galaxy power spectra are limited by the perfect degeneracy between the LPnG parameter $f_{\rm NL}$ and the bias parameter $b_ϕ$ which encodes the response of galaxy clustering to a change in the amplitude of primordial curvature fluctuations. For galaxies observed by galaxy surveys, the relation between $b_ϕ$ and the galaxy bias $b_{g}$ is poorly understood and differs significantly from the universal mass function ansatz. I…

LLM-empowered Dynamic Prompt Routing for Vision-Language Models Tuning under Long-Tailed Distributions
Yongju Jia, Jiarui Ma, Xiangxian Li, Baiqiao Zhang, Xianhui Cao, Juan Liu, Yulong Bian
https://arxiv.org/abs/2508.15688

LLM-empowered Dynamic Prompt Routing for Vision-Language Models Tuning under Long-Tailed Distributions
Pre-trained vision-language models (VLMs), such as CLIP, have demonstrated impressive capability in visual tasks, but their fine-tuning often suffers from bias in class-imbalanced scene. Recent works have introduced large language models (LLMs) to enhance VLM fine-tuning with supplementing semantic information. However, they often overlook inherent class imbalance in VLMs' pre-training, which may lead to bias accumulation in downstream tasks. To address this problem, this paper proposes a Multi…

A Comprehensive Evaluation of the Sensitivity of Density-Ratio Estimation Based Fairness Measurement in Regression
Abdalwahab Almajed, Maryam Tabar, Peyman Najafirad
https://arxiv.org/abs/2508.14576

A Comprehensive Evaluation of the Sensitivity of Density-Ratio Estimation Based Fairness Measurement in Regression
The prevalence of algorithmic bias in Machine Learning (ML)-driven approaches has inspired growing research on measuring and mitigating bias in the ML domain. Accordingly, prior research studied how to measure fairness in regression which is a complex problem. In particular, recent research proposed to formulate it as a density-ratio estimation problem and relied on a Logistic Regression-driven probabilistic classifier-based approach to solve it. However, there are several other methods to esti…

How built environment shapes cycling experience: A multi-scale review in historical urban contexts
Haining Ding, Chenxi Wang, Michal Gath-Morad
https://arxiv.org/abs/2509.15657 …

How built environment shapes cycling experience: A multi-scale review in historical urban contexts
Understanding how built environments shape human experience is central to designing sustainable cities. Cycling provides a critical case: it delivers health and environmental benefits, yet its uptake depends strongly on the experience of cycling rather than infrastructure alone. Research on this relationship has grown rapidly but remains fragmented across disciplines and scales, and has concentrated on network-level analyses of routes and connectivity. This bias is especially problematic in his…

PC-Sampler: Position-Aware Calibration of Decoding Bias in Masked Diffusion Models
Pengcheng Huang, Shuhao Liu, Zhenghao Liu, Yukun Yan, Shuo Wang, Zulong Chen, Tong Xiao
https://arxiv.org/abs/2508.13021

PC-Sampler: Position-Aware Calibration of Decoding Bias in Masked Diffusion Models
Recent advances in masked diffusion models (MDMs) have established them as powerful non-autoregressive alternatives for sequence generation. Nevertheless, our preliminary experiments reveal that the generation quality of MDMs is still highly sensitive to the choice of decoding strategy. In particular, widely adopted uncertainty-based samplers suffer from two key limitations: a lack of global trajectory control and a pronounced bias toward trivial tokens in the early stages of decoding. These sh…

BSI veröffentlicht Whitepaper zum Bias in der KI
Wenn sich KI-Systeme nicht so verhalten, wie erwünscht, stecken oft Verzerrungen in den Daten dahinter. Das BSI erklärt, wie man den Bias im System erkennt.
https://www.

Design of a Gm-C Dynamic Amplifier with High Linearity and High Temperature and Power Supply Voltage Stability
Jinkun Yang, Pengbin Xu
https://arxiv.org/abs/2508.14637 https://

Design of a Gm-C Dynamic Amplifier with High Linearity and High Temperature and Power Supply Voltage Stability
This paper presents a Gm-C dynamic amplifier with high linearity and high temperature and power supply voltage stability. The main part of the amplifier employs two asymmetric differential pairs to enhance transconductance linearity. The amplifier maintains a nearly constant gain within a differential input range of -40 mV to 40 mV, and achieves a total harmonic distortion (THD) of 70.5 dB. The bias part of the amplifier adopts a constant-gm bias circuit, which improves the temperature and supp…

Fairness for the People, by the People: Minority Collective Action
Omri Ben-Dov, Samira Samadi, Amartya Sanyal, Alexandru \c{T}ifrea
https://arxiv.org/abs/2508.15374 https://

Fairness for the People, by the People: Minority Collective Action
Machine learning models often preserve biases present in training data, leading to unfair treatment of certain minority groups. Despite an array of existing firm-side bias mitigation techniques, they typically incur utility costs and require organizational buy-in. Recognizing that many models rely on user-contributed data, end-users can induce fairness through the framework of Algorithmic Collective Action, where a coordinated minority group strategically relabels its own data to enhance fairne…

Who Gets the Mic? Investigating Gender Bias in the Speaker Assignment of a Speech-LLM
Dariia Puhach, Amir H. Payberah, \'Eva Sz\'ekely
https://arxiv.org/abs/2508.13603 h…

Who Gets the Mic? Investigating Gender Bias in the Speaker Assignment of a Speech-LLM
Similar to text-based Large Language Models (LLMs), Speech-LLMs exhibit emergent abilities and context awareness. However, whether these similarities extend to gender bias remains an open question. This study proposes a methodology leveraging speaker assignment as an analytic tool for bias investigation. Unlike text-based models, which encode gendered associations implicitly, Speech-LLMs must produce a gendered voice, making speaker selection an explicit bias cue. We evaluate Bark, a Text-to-Sp…

High-Frequency First: A Two-Stage Approach for Improving Image INR
Sumit Kumar Dam, Mrityunjoy Gain, Eui-Nam Huh, Choong Seon Hong
https://arxiv.org/abs/2508.15582 https://

High-Frequency First: A Two-Stage Approach for Improving Image INR
Implicit Neural Representations (INRs) have emerged as a powerful alternative to traditional pixel-based formats by modeling images as continuous functions over spatial coordinates. A key challenge, however, lies in the spectral bias of neural networks, which tend to favor low-frequency components while struggling to capture high-frequency (HF) details such as sharp edges and fine textures. While prior approaches have addressed this limitation through architectural modifications or specialized …

Combating Homelessness Stigma with LLMs: A New Multi-Modal Dataset for Bias Detection
Jonathan A. Karr Jr., Benjamin F. Herbst, Ting Hua, Matthew Hauenstein, Georgina Curto, Nitesh V. Chawla
https://arxiv.org/abs/2508.13187

Combating Homelessness Stigma with LLMs: A New Multi-Modal Dataset for Bias Detection
Homelessness is a persistent social challenge, impacting millions worldwide. Over 770,000 people experienced homelessness in the U.S. in 2024. Social stigmatization is a significant barrier to alleviation, shifting public perception, and influencing policymaking. Given that online and city council discourse reflect and influence part of public opinion, it provides valuable insights to identify and track social biases. This research contributes to alleviating homelessness by acting on public opi…

Sequential Confirmatory Factor Analysis: A Novel Approach to Latent Variable Measurement
Zachary Esses Johnson
https://arxiv.org/abs/2508.15611 https://arx…

Sequential Confirmatory Factor Analysis: A Novel Approach to Latent Variable Measurement
Factor score estimation in small sample sizes often encounters parameter bias and convergence failures when constructing hierarchical national/sub-national indices. This paper proposes a novel method for hierarchical factor analysis called "sequential Confirmatory Factor Analysis". Instead of estimating multiple levels of factors at the same time, this approach calculates factor scores sequentially from the lowest to highest levels. This sequential estimation keeps the original sample size in e…

REFER: Mitigating Bias in Opinion Summarisation via Frequency Framed Prompting
Nannan Huang, Haytham M. Fayek, Xiuzhen Zhang
https://arxiv.org/abs/2509.15723 https://

REFER: Mitigating Bias in Opinion Summarisation via Frequency Framed Prompting
Individuals express diverse opinions, a fair summary should represent these viewpoints comprehensively. Previous research on fairness in opinion summarisation using large language models (LLMs) relied on hyperparameter tuning or providing ground truth distributional information in prompts. However, these methods face practical limitations: end-users rarely modify default model parameters, and accurate distributional information is often unavailable. Building upon cognitive science research demo…

Noch einige der zuletzt hier besonders häufig geteilten #News:
Trumps anti-woke-Verordnung: Meta arbeitet mit rechtem Influencer

When Audio and Text Disagree: Revealing Text Bias in Large Audio-Language Models
Cheng Wang, Gelei Deng, Xianglin Yang, Han Qiu, Tianwei Zhang
https://arxiv.org/abs/2508.15407 h…

When Audio and Text Disagree: Revealing Text Bias in Large Audio-Language Models
Large Audio-Language Models (LALMs) are enhanced with audio perception capabilities, enabling them to effectively process and understand multimodal inputs that combine audio and text. However, their performance in handling conflicting information between audio and text modalities remains largely unexamined. This paper introduces MCR-BENCH, the first comprehensive benchmark specifically designed to evaluate how LALMs prioritize information when presented with inconsistent audio-text pairs. Throu…

Estimating systematic errors in Bayesian inversion using transport maps
Maren Casfor, Philipp Trunschke, Sebastian Heidenreich, Nando Hegemann
https://arxiv.org/abs/2509.16116 h…

Estimating systematic errors in Bayesian inversion using transport maps
In indirect measurements, the measurand is determined by solving an inverse problem which requires a model of the measurement process. Such models are often approximations and introduce systematic errors leading to a bias of the posterior distribution in Bayesian inversion. We propose a unified framework that combines transport maps from a reference distribution to the posterior distribution with the model error approach. This leads to an adaptive algorithm that jointly estimates the posterior …

A Race Bias Free Face Aging Model for Reliable Kinship Verification
Ali Nazari, Bardiya Kariminia, Mohsen Ebrahimi Moghaddam
https://arxiv.org/abs/2509.15177 https://

A Race Bias Free Face Aging Model for Reliable Kinship Verification
The age gap in kinship verification addresses the time difference between the photos of the parent and the child. Moreover, their same-age photos are often unavailable, and face aging models are racially biased, which impacts the likeness of photos. Therefore, we propose a face aging GAN model, RA-GAN, consisting of two new modules, RACEpSp and a feature mixer, to produce racially unbiased images. The unbiased synthesized photos are used in kinship verification to investigate the results of ver…

Adversarial Hospital-Invariant Feature Learning for WSI Patch Classification
Mengliang Zhang, Jacob M. Luber
https://arxiv.org/abs/2508.14779 https://arxiv…

Adversarial Hospital-Invariant Feature Learning for WSI Patch Classification
Pathology foundation models (PFMs) have demonstrated remarkable potential in whole-slide image (WSI) diagnosis. However, pathology images from different hospitals often vary due to differences in scanning hardware and preprocessing styles, which may lead PFMs to inadvertently learn hospital-specific features, posing risks for clinical deployment. In this work, we present the first systematic study of domain bias in PFMs arising from hospital source characteristics. Specifically, we (1) construc…

Mitigating Strategy Preference Bias in Emotional Support Conversation via Uncertainty Estimations
Yougen Zhou, Qin Chen, Ningning Zhou, Jie Zhou, Xingjiao Wu, Liang He
https://arxiv.org/abs/2509.12661 …

Mitigating Strategy Preference Bias in Emotional Support Conversation via Uncertainty Estimations
Emotional support conversation (ESC) aims to alleviate distress through empathetic dialogue, yet large language models (LLMs) face persistent challenges in delivering effective ESC due to low accuracy in strategy planning. Moreover, there is a considerable preference bias towards specific strategies. Prior methods using fine-tuned strategy planners have shown potential in reducing such bias, while the underlying causes of the preference bias in LLMs have not well been studied. To address these …

Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA
Kaiwei Zhang, Qi Jia, Zijian Chen, Wei Sun, Xiangyang Zhu, Chunyi Li, Dandan Zhu, Guangtao Zhai
https://arxiv.org/abs/2508.13743

Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA
Large language models (LLMs), while increasingly used in domains requiring factual rigor, often display a troubling behavior: sycophancy, the tendency to align with user beliefs regardless of correctness. This tendency is reinforced by preference-based alignment techniques that optimize for user satisfaction but can undermine truthfulness. While relatively benign in casual dialogue, sycophancy poses serious risks in high-stakes settings such as scientific question answering (QA), where model ou…

Measuring Gender Bias in Job Title Matching for Grammatical Gender Languages
Laura Garc\'ia-Sardi\~na, Hermenegildo Fabregat, Daniel Deniz, Rabih Zbib
https://arxiv.org/abs/2509.13803

Measuring Gender Bias in Job Title Matching for Grammatical Gender Languages
This work sets the ground for studying how explicit grammatical gender assignment in job titles can affect the results of automatic job ranking systems. We propose the usage of metrics for ranking comparison controlling for gender to evaluate gender bias in job title ranking systems, in particular RBO (Rank-Biased Overlap). We generate and share test sets for a job title matching task in four grammatical gender languages, including occupations in masculine and feminine form and annotated by gen…

Inductive Bias Extraction and Matching for LLM Prompts
Christian M. Angel, Francis Ferraro
https://arxiv.org/abs/2508.10295 https://arxiv.org/pdf/2508.1029…

Inductive Bias Extraction and Matching for LLM Prompts
The active research topic of prompt engineering makes it evident that LLMs are sensitive to small changes in prompt wording. A portion of this can be ascribed to the inductive bias that is present in the LLM. By using an LLM's output as a portion of its prompt, we can more easily create satisfactory wording for prompts. This has the effect of creating a prompt that matches the inductive bias in model. Empirically, we show that using this Inductive Bias Extraction and Matching strategy improves …

BIPOLAR: Polarization-based granular framework for LLM bias evaluation
Martin Pavl\'i\v{c}ek, Tom\'a\v{s} Filip, Petr Sos\'ik
https://arxiv.org/abs/2508.11061 https:…

BIPOLAR: Polarization-based granular framework for LLM bias evaluation
Large language models (LLMs) are known to exhibit biases in downstream tasks, especially when dealing with sensitive topics such as political discourse, gender identity, ethnic relations, or national stereotypes. Although significant progress has been made in bias detection and mitigation techniques, certain challenges remain underexplored. This study proposes a reusable, granular, and topic-agnostic framework to evaluate polarisation-related biases in LLM (both open-source and closed-source). …

Fair-GPTQ: Bias-Aware Quantization for Large Language Models
Irina Proskurina, Guillaume Metzler, Julien Velcin
https://arxiv.org/abs/2509.15206 https://ar…

Fair-GPTQ: Bias-Aware Quantization for Large Language Models
High memory demands of generative language models have drawn attention to quantization, which reduces computational cost, memory usage, and latency by mapping model weights to lower-precision integers. Approaches such as GPTQ effectively minimize input-weight product errors during quantization; however, recent empirical studies show that they can increase biased outputs and degrade performance on fairness benchmarks, and it remains unclear which specific weights cause this issue. In this work, …

Analysing Moral Bias in Finetuned LLMs through Mechanistic Interpretability
Bianca Raimondi, Daniela Dalbagno, Maurizio Gabbrielli
https://arxiv.org/abs/2510.12229 https://

Analysing Moral Bias in Finetuned LLMs through Mechanistic Interpretability
Large language models (LLMs) have been shown to internalize human-like biases during finetuning, yet the mechanisms by which these biases manifest remain unclear. In this work, we investigated whether the well-known Knobe effect, a moral bias in intentionality judgements, emerges in finetuned LLMs and whether it can be traced back to specific components of the model. We conducted a Layer-Patching analysis across 3 open-weights LLMs and demonstrated that the bias is not only learned during finet…