论文检索

Disa Sariola, Patrick Button, Aron Culotta, Nicholas Mattei

Classifiers trained on historical data are deployed in the real world to automate decisions from hiring to loan issuance. Judging the fairness and efficiency of these systems, and their human counterparts, is a complex and important topic studied across both computational and social sciences. One common way to address bias in classifiers is to resample the training data to offset distributional disparities. In the hiring domain, where results may vary by a protected class, many interventions from the literature equalize the hiring rate within the training set to alleviate bias. While simple and seemingly effective, these methods have typically only been evaluated using data obtained through convenience samples, e.g., data from a real-world hiring process, introducing selection and label bias. In the social and health sciences, audit studies, in which fictitious "testers" (resumes) are sent to subjects (job openings) in a randomized control trial, provide high-quality data that support rigorous estimates of discrimination by controlling for confounding factors. We investigate how data from audit studies can be used to improve our ability to both train and evaluate automated hiring algorithms. Specifically, we use data from a large audit study of age discrimination in hiring to test common resampling methods from the fair machine learning literature. We find that audit data of real-world hiring reveals cases where equalizing base rates across classes appears to achieve parity using traditional measures, but in fact has an absolute ~10% disparity when measured appropriately. We also show that corrections based on individual treatment effect estimation methods combined with audit study data can overcome these issues, underscoring the need for rigorous data collection in fairness research.

Depanshu Sani, Mehar Khurana, Saket Anand

Animal re-identification (Re-ID) has recently gained substantial attention in the AI research community due to its high impact on biodiversity monitoring and unique research challenges arising from environmental factors. The subtle distinguishing patterns like stripes or spots, handling new species and the inherent open-set nature make the problem even harder. To address these complexities, foundation models trained on labeled, large-scale and multi-species animal Re-ID datasets have recently been introduced to enable zero-shot Re-ID. However, our benchmarking reveals significant gaps in their zero-shot Re-ID performance for both known and unknown species. While this highlights the need for collecting labeled data in new domains, exhaustive annotation for Re-ID is laborious and requires domain expertise. Our analyses also show that existing unsupervised (USL) and active learning (AL) Re-ID methods underperform for animal Re-ID. To address these limitations, we introduce a novel AL Re-ID framework that leverages complementary clustering methods to uncover and target structurally ambiguous regions in the embedding space for mining pairs of samples that are both informative and broadly representative. Oracle feedback on these pairs, in the form of must-link and cannot-link constraints, facilitates a simple annotation interface, which naturally integrates with existing USL methods through our proposed constrained clustering refinement algorithm. Through extensive experiments, we demonstrate that, by utilizing only 0.033% of all possible annotations, our approach consistently outperforms existing foundational, USL and AL baselines. Specifically, we report an average improvement of 10.49%, 11.19% and 3.99% (mAP) on 13 wildlife datasets over foundational, USL and AL methods, respectively, while attaining state-of-the-art performance on each dataset. Furthermore, we also show an improvement of 11.09%, 8.2% and 2.06% (AUC ROC) for unknown individuals in an open-world setting. We also present results on 2 publicly available person Re-ID datasets, showing average gains of 7.96% and 2.86% (mAP) over existing USL and AL Re-ID methods.

Ahmed Sabir, Markus Kängsepp, Rajesh Sharma

The increased use of large language models (LLMs) in sensitive domains leads to growing interest in how their confidence scores correspond to fairness and bias. This study examines the alignment between LLM-predicted confidence and human-annotated bias judgments. Focusing on gender bias, the research investigates probability confidence calibration in contexts involving gendered pronoun resolution. The goal is to evaluate if calibration metrics based on predicted confidence scores effectively capture fairness-related disparities in LLMs. The results show that, among the six state-of-the-art models, Gemma-2 demonstrates the worst calibration according to the gender bias benchmark. The primary contribution of this work is a fairness-aware evaluation of LLMs confidence calibration, offering guidance for ethical deployment. In addition, we introduce a new calibration metric, Gender-ECE, designed to measure gender disparities in resolution tasks.

Weilin Ruan, Xilin Dang, Ziyu Zhou, Sisuo Lyu, Yuxuan Liang

Traffic prediction serves as a cornerstone of modern intelligent transportation systems and the critical task of spatio-temporal forecasting. Although advanced Spatio-temporal Graph Neural Networks (STGNNs) and pre-trained models have made significant progress in traffic prediction, two critical challenges persist: (i) limited contextual capacity when handling complex spatio-temporal dependencies, and (ii) low predictability at fine-grained spatio-temporal points caused by heterogeneous patterns. Inspired by Retrieval-Augmented Generation (RAG), we propose RAST, a universal framework that integrates retrieval-augmented mechanisms with spatio-temporal modeling to address these challenges. Our framework consists of three key designs: 1) Decoupled Encoder and Query Generator to capture decoupled spatial and temporal features and construct a fusion query via residual fusion; 2) Spatio-temporal Retrieval Store and Retrievers to maintain and retrieve vectorized fine-grained patterns; and 3) Universal Backbone Predictor that flexibly accommodates pre-trained STGNNs or simple MLP predictors. Extensive experiments on 6 real-world traffic networks, including large-scale datasets, demonstrate that RAST achieves superior performance while maintaining computational efficiency.

Oscar J. Romero, Anthony Tomasic, Elizabeth J. Carter, John Zimmerman, Aaron Steinfeld

Navigating new indoor spaces and interacting with the environment presents many challenges for people who are blind or have low vision (BLV). To address these challenges, we prototyped a smartphone-based conversational assistant that helps BLV people navigate and interact with their environment. The prototype utilizes a cognitive architecture to integrate three different technologies: (i) augmented-reality spatial anchors for high-precision localization and access to static information about the environment; (ii) real-time object/people detection for information about the environment and obstacle avoidance; and (iii) a conversational agent}that uses large language models (LLMs) for information extraction, conversational interaction, and turn-by-turn navigation. We assess the impact of different technologies on human performance by measuring user task time and errors. We found that conversational interaction holistically integrates the different technologies to deliver a better user experience while significantly reducing task completion time.

Michael Richards, Danny Cowser, Daniel Nielson, Jacob W. Crandall

Institutions are key to creating societies that are efficient, fair, and benevolent. Despite their importance, the complexities of human (networked) societies make it difficult to understand how formal institutions form and how they shape human communities. Artificial intelligence (AI) can potentially raise understanding in this regard. Thus, in this paper, we present a simulation model utilizing AI agents to simulate networked societies that contain formal institutions. We then observe the outputs of the resulting model under different societal conditions and formal institutions, and (where applicable) compare and contrast these outputs with political and economic theories. Our model outputs (a) address how inequality impacts societal prosperity, (b) illuminate how institutions can potentially impact poverty, and (c) give insights into the attributes of formal institutions that individuals are inclined to support. These and future simulation models can potentially inform how AI can support the design and development of institutions that facilitate healthier communities and nations.

Mohammad Zia Ur Rehman, Sai Kartheek Reddy Kasu, Shashivardhan Reddy Koppula, Sai Rithwik Reddy Chirra, Shwetank Shekhar Singh, Nagendra Kumar

Hate speech detection on social media faces challenges in both accuracy and explainability, especially for underexplored Indic languages. We propose a novel explainability-guided training framework, X-MuTeST (eXplainable Multilingual haTe Speech deTection), for hate speech detection that combines high-level semantic reasoning from large language models (LLMs) with traditional attention-enhancing techniques. We extend this research to Hindi and Telugu alongside English by providing benchmark human-annotated rationales for each word to justify the assigned class label. The X-MuTeST explainability method computes the difference between the prediction probabilities of the original text and those of unigrams, bigrams, and trigrams. Final explanations are computed as the union between LLM explanations and X-MuTeST explanations. We show that leveraging human rationales during training enhances both classification performance and the model’s explainability. Moreover, combining human rationales with our explainability method to refine the model’s attention yields further improvements. We evaluate explainability using Plausibility metrics such as Token-F1 and IOU-F1, and Faithfulness metrics such as Comprehensiveness and Sufficiency. By focusing on under-resourced languages, our work advances hate speech detection across diverse linguistic contexts. Our dataset includes token-level rationale annotations for 6,004 Hindi, 4,492 Telugu, and 6,334 English samples.

Hasibur Rahman, Hanan Salam

Large language models (LLMs) increasingly shape interpersonal and societal decision-making, yet their ability to navigate explicit conflicts between legitimate cultural values remains underexplored. Existing benchmarks focus on cultural knowledge (CulturalBench), value inference (WorldValuesBench), or single-axis bias (CDEval), but none assess how LLMs adjudicate when multiple cultural frameworks directly clash. We introduce CCD-Bench (Culture-Conflict Decision Benchmark), a benchmark for evaluating LLM decision-making under cross-cultural value conflict. CCD-Bench contains 2,182 open-ended dilemmas across seven domains, each with ten anonymized response options aligned with the ten GLOBE cultural clusters spanning 62 societies. Using a Stratified Latin Square design, we evaluate 17 leading LLMs and find clear biases: models favor Nordic Europe (20.2%) and Germanic Europe (12.4%), while Eastern Europe and Middle East & North Africa responses are least preferred (≈5–6%). Although 87.9% of model rationales reference multiple cultural dimensions, this pluralism is shallow, dominated by Future and Performance Orientation, with limited attention to Assertiveness or Gender Egalitarianism (

Elinor Poole-Dayan, Deb Roy, Jad Kabbara

While state-of-the-art large language models (LLMs) have shown impressive performance on many tasks, systematically evaluating undesirable behaviors of these models remains critical. In this work, we investigate how the quality of LLM responses changes in terms of information accuracy, truthfulness, and refusals depending on three user traits: English proficiency, education level, and country of origin. We present extensive experimentation on three state-of-the-art LLMs and two different datasets targeting truthfulness and factuality. Our findings suggest that undesirable behaviors in state-of-the-art LLMs occur disproportionately more for users with lower English proficiency, of lower education status, and originating from outside the US, rendering these models unreliable sources of information towards their most vulnerable users.

Markus B. Pettersson, Connor Thomas Jerzak, Adel Daoud

Machine learning models trained on Earth observation data, such as satellite imagery, have demonstrated significant promise in predicting household-level wealth indices, enabling the creation of high-resolution wealth maps that can be leveraged across multiple causal trials while addressing chronic data scarcity in global development research. However, because standard training objectives prioritize overall predictive accuracy, these predictions often suffer from shrinkage toward the mean, leading to attenuated estimates of causal treatment effects and limiting their utility in policy evaluations. Existing debiasing methods, such as Prediction-Powered Inference (PPI), can handle this attenuation bias but require additional fresh ground-truth data at the downstream stage of causal inference, which restricts their applicability in data-scarce environments. We introduce and evaluate two post-hoc correction methods—Linear Calibration Correction (LCC) and a Tweedie's correction approach—that substantially reduce shrinkage-induced prediction bias without relying on newly collected labeled data. LCC applies a simple linear transformation estimated on a held-out calibration split; Tweedie's method locally de-shrink predictions using density score estimates and a noise scale learned upstream. We provide practical diagnostics for when a correction is warranted and discuss practical limitations. Across analytical results, simulations, and experiments with Demographic and Health Surveys (DHS) data, both approaches reduce attenuation; Tweedie's correction yields nearly unbiased treatment-effect estimates, enabling a "one map, many trials" paradigm. Although we demonstrate on EO-ML wealth mapping, the methods are not geospatial-specific: they apply to any setting where imputed outcomes are reused downstream (e.g., pollution indices, population density, or LLM-derived indicators).

Zeel B Patel, Vinayak Rana, Nipun Batra

Air pollution is a leading global health threat, yet many developing countries lack the dense monitoring infrastructure needed for accurate exposure assessment and informed policy. Optimal Sensor Placement (OSP) is a foundational challenge in expanding monitoring capacity. While mutual information (MI) offers a principled criterion for selecting informative sensor locations, its computational cost grows with both the number of placements and the density of the candidate grid. We present a scalable, continuous optimization framework that treats sensor coordinates as differentiable parameters and directly maximizes MI. Unlike standard approaches, our method is computationally efficient—its runtime is independent of both the number of placements and the size of the search grid—making MI-based acquisition feasible over large spatial domains. On a continental-scale PM2.5 dataset, our method outperforms random placement and the widely-used Maximum Predictive Variance heuristic. In a focused regional study, it approaches the performance of greedy MI while being orders of magnitude faster. Our framework enables practical, information-theoretic sensor placement for real-world environmental monitoring.

Joan Nwatu, Longju Bai, Oana Ignat, Rada Mihalcea

Culture shapes the objects people use and for what purposes, yet mainstream Vision-Language (VL) datasets frequently exhibit cultural biases, disproportionately favoring higher-income, Western contexts. This imbalance reduces model generalizability and perpetuates performance disparities, especially impacting lower-income and non-Western communities. To address these disparities, we propose a novel function-centric framework that categorizes objects by the functions they fulfill, across diverse cultural and economic contexts. We implement this framework by creating the Culture Affordance Atlas, a re-annotated and culturally grounded restructuring of the Dollar Street dataset spanning 46 functions and 288 objects. Through extensive empirical analyses using the CLIP model, we demonstrate that function-centric labels substantially reduce socioeconomic performance gaps between high and low-income groups by a median of 6 pp (statistically significant), improving model effectiveness for lower income contexts. Furthermore, our analyses reveals numerous culturally essential objects that are frequently overlooked in prominent VL datasets. Our contributions offer a scalable pathway toward building inclusive VL datasets and equitable AI systems.

Shahriar Noroozizadeh, Sayantan Kumar, Jeremy C. Weiss

Clinical case reports encode temporal patient trajectories that are often underexploited by traditional machine learning methods relying on structured data. In this work, we introduce the forecasting problem from textual time series, where timestamped clinical findings—extracted via an LLM-assisted annotation pipeline—serve as the primary input for prediction. We systematically evaluate a diverse suite of models, including fine-tuned decoder-based large language models and encoder-based transformers, on tasks of event occurrence prediction, temporal ordering, and survival analysis. Our experiments reveal that encoder-based models consistently achieve higher F1 scores and superior temporal concordance for short- and long-horizon event forecasting, while fine-tuned masking approaches enhance ranking performance. In contrast, instruction-tuned decoder models demonstrate a relative advantage in survival analysis, especially in early prognosis settings. Our sensitivity analyses further demonstrate the importance of time ordering, which requires clinical time series construction, as compared to text ordering, the format of the text inputs that LLMs are classically trained on. This highlights the additional benefit that can be ascertained from time-ordered corpora, with implications for temporal tasks in the era of widespread LLM use.

Terrence Neumann, Maria De-Arteaga, Sina Fazelpour

The emergent capabilities of large language models (LLMs) have prompted interest in using them as surrogates for human subjects in opinion surveys. However, prior evaluations of LLM-based opinion simulation have relied heavily on costly, domain-specific survey data, and mixed empirical results leave their reliability in question. To enable cost-effective, early-stage evaluation, we introduce a quality control assessment designed to test the viability of LLM-simulated opinions on Likert-scale tasks without requiring large-scale human data for validation. This assessment comprises two key tests: logical consistency and alignment with stakeholder expectations, offering a low-cost, domain-adaptable validation tool. We apply our quality control assessment to an opinion simulation task relevant to AI-assisted content moderation and fact-checking workflows---a socially impactful use case---and evaluate nine LLMs using a baseline prompt engineering method (backstory prompting), as well as fine-tuning and in-context learning variants. None of the models or methods pass the full assessment, revealing several failure modes. We conclude with a discussion of the risk management implications and release TopicMisinfo, a benchmark dataset with paired human and LLM annotations simulated by various models and approaches, to support future research.

Soumyaroop Nandi, Prem Natarajan

Scientific image manipulation in biomedical publications poses a growing threat to research integrity and reproducibility. Unlike natural image forensics, biomedical forgery detection is uniquely challenging due to domain-specific artifacts, complex textures, and unstructured figure layouts. We present the first vision-language guided framework for both generating and detecting biomedical image forgeries. By combining diffusion-based synthesis with vision-language prompting, our method enables realistic and semantically controlled manipulations—including duplication, splicing, and region removal—across diverse biomedical modalities. We introduce Rescind, a large-scale benchmark featuring fine-grained annotations and modality-specific splits, and propose Integscan, a structured state-space modeling framework that integrates attention-enhanced visual encoding with prompt-conditioned semantic alignment for precise forgery localization. To ensure semantic fidelity, we incorporate a VLM-based verification loop that filters generated forgeries based on consistency with intended prompts. Extensive experiments on Rescind and existing benchmarks demonstrate that Integscan achieves state-of-the-art performance in both detection and localization, establishing a strong foundation for automated scientific integrity analysis.

Sujay Nair, Evan Austen Coleman, Sherrie Wang, Elsa Olivetti

Minerals play a critical role in the advanced energy technologies necessary for decarbonization, but characterizing mineral deposits hidden underground remains costly and challenging. Inspired by recent progress in generative modeling, we develop a learning method which infers the locations of minerals by masking and infilling geospatial maps of resource availability. We demonstrate this technique using mineral data for the conterminous United States, and train performant models, with the best achieving Dice coefficients of 0.31 ± 0.01 and recalls of 0.22 ± 0.02 on test data at 1×1 sq mi spatial resolution. One major advantage of our approach is that it can easily incorporate auxiliary data sources for prediction which may be more abundant than mineral data. We highlight the capabilities of our model by adding input layers derived from geophysical sources, along with a nation-wide ground survey of soils originally intended for agronomic purposes. We find that employing such auxiliary features can improve inference performance, while also enabling model evaluation in regions with no recorded minerals.

Manuel Mosquera, Melissa Verónica Robles, Johan R. Portela, Rubén Manrique

Low-resource machine translation remains a significant challenge for large language models (LLMs), which often lack exposure to these languages during pretraining and have limited parallel data for fine-tuning. We propose a novel approach that enhances translation for low-resource languages by integrating an external dictionary tool and training models end-to-end using reinforcement learning, in addition to supervised fine-tuning. Focusing on the Spanish–Wayuunaiki language pair, we frame translation as a tool-augmented decision-making problem in which the model can selectively consult a bilingual dictionary during generation. Our method combines supervised instruction tuning with Group Relative Policy Optimization (GRPO), enabling the model to learn both when and how to use the tool effectively. BLEU similarity scores are used as rewards to guide this learning process. Preliminary results show that our tool-augmented models achieve up to +3.37 BLEU improvement over previous work and an 18% relative gain compared to a supervised baseline without dictionary access, on the Spanish–Wayuunaiki test set from the AmericasNLP 2025 Shared Task. We also conduct ablation studies to assess the effects of model architecture and training strategy, comparing Qwen2.5-0.5B-Instruct with other models such as LLaMA and a prior NLLB-based system. These findings highlight the promise of combining LLMs with external tools and the role of reinforcement learning in improving translation quality in low-resource language settings.

Xiaoyue Mi, Fan Tang, You Wu, Juan Cao, Peng Li, Yang Liu

Personalized concept generation by tuning diffusion models with a few images raises potential legal and ethical concerns regarding privacy and intellectual property rights. Researchers attempt to prevent malicious personalization using adversarial perturbations. However, previous efforts have mainly focused on the effectiveness of protection while neglecting the visibility of perturbations. They utilize global adversarial perturbations, which introduce noticeable alterations to original images and significantly degrade visual quality. In this work, we propose the Visual-Friendly Concept Protection (VCPro) framework, which prioritizes the protection of key concepts chosen by the image owner through adversarial perturbations with lower perceptibility. To ensure these perturbations are as inconspicuous as possible, we introduce a relaxed optimization objective to identify the least perceptible yet effective adversarial perturbations, solved using the Lagrangian multiplier method. Qualitative and quantitative experiments validate that VCPro achieves a better trade-off between the visibility of perturbations and protection effectiveness, effectively prioritizing the protection of target concepts in images with less perceptible perturbations.

Yi Mao, Andrew Perrault

Municipal inspections are an important part of maintaining the quality of goods and services. In this paper, we approach the problem of intelligently scheduling service inspections to maximize their impact, using the case of food establishment inspections in Chicago as a case study. The Chicago Department of Public Health (CDPH) inspects thousands of establishments each year, with a substantial fail rate (over 3,000 failed inspection reports in 2023). To balance the objectives of ensuring adherence to guidelines, minimizing disruption to establishments, and minimizing inspection costs, CDPH assigns each establishment an inspection window every year and guarantees that they will be inspected exactly once during that window. Meanwhile, CDPH also promises surprise public health inspections for unexpected food safety emergencies or complaints. These constraints create a challenge for a restless multi-armed bandit (RMAB) approach, for which there are no existing methods. We develop an extension to Whittle index-based systems for RMABs that can guarantee action window constraints and frequencies, and furthermore can be leveraged to optimize action window assignments themselves. Briefly, we combine MDP reformulation and integer programming-based lookahead to maximize the impact of inspections subject to constraints. A neural network-based supervised learning model is developed to model state transitions of real Chicago establishments using public CDPH inspection records, which demonstrates 10% AUC improvements compared with directly predicting establishments' failures. Our experiments not only show up to 24% (in simulation) or 33% (on real data) objective improvements resulting from our approach and robustness to surprise inspections, but also give insight into the impact of scheduling constraints.

Thomas Manzini, Priyankari Perali, Raisa Karnik, Robin R. Murphy

This paper presents the largest known benchmark dataset for road damage assessment and road alignment, and provides 18 baseline models trained on the CRASAR-U-DRIODs dataset's post-disaster small uncrewed aerial systems (sUAS) imagery from 10 federally declared disasters, addressing three challenges within prior post-disaster road damage assessment datasets. While prior disaster road damage assessment datasets exist, there is no current state of practice, as prior public datasets have either been small-scale or reliant on low-resolution imagery insufficient for detecting phenomena of interest to emergency managers. Further, while machine learning (ML) systems have been developed for this task previously, none are known to have been operationally validated. These limitations are overcome in this work through the labeling of 657.25km of roads according to a 10-class labeling schema, followed by training and deploying ML models during the operational response to Hurricanes Debby and Helene in 2024. Motivated by observed road line misalignment in practice, 9,184 road line adjustments were provided for spatial alignment of a priori road lines, as it was found that when the 18 baseline models are deployed against real-world misaligned road lines, model performance degraded on average by 5.596% Macro IoU. If spatial alignment is not considered, approximately 8% (11km) of adverse conditions on road lines will be labeled incorrectly, with approximately 9% (59km) of road lines misaligned off the actual road. These dynamics are gaps that should be addressed by the ML, CV, and robotics communities to enable more effective and informed decision-making during disasters.