Research Papers

AI expansion has accelerated workplace adoption of new technologies. Yet, it is unclear whether and how knowledge workers are supported and trained to safely use AI. Inadequate training may lead to unrealized benefits if workers abandon tools, or perpetuate biases if workers misinterpret AI-based outcomes. In a workshop with 39 workers from 26 countries specializing in human resources, labor law, standards creation, and worker training, we explored questions and ideas they had about safely adopting AI. We held 17 follow-up interviews to further investigate what skills and training knowledge workers need to achieve safe and effective AI in practice. We synthesize nine training topics participants surfaced for knowledge workers related to challenges around understanding what AI is, misinterpreting outcomes, exacerbating biases, and worker rights. We reflect how these training topics might be addressed under different contexts, imagine HCI research prototypes as potential training tools, and consider ways to ensure training does not perpetuate harmful values.

The remarkable capabilities of Large Language Models (LLMs) make them increasingly compelling for adoption in real-world healthcare applications. However, the risks associated with using LLMs in medical applications have not been systematically characterized. We propose using five key principles for safe and trustworthy medical AI: Truthfulness, Resilience, Fairness, Robustness, and Privacy, along with ten specific aspects. Under this comprehensive framework, we introduce a novel MedGuard benchmark with 1,000 expert-verified questions. Our evaluation of 11 commonly used LLMs shows that the current language models, regardless of their safety alignment mechanisms, generally perform poorly on most of our benchmarks, particularly when compared to the high performance of human physicians. Despite recent reports indicate that advanced LLMs like ChatGPT can match or even exceed human performance in various medical tasks, this study underscores a significant safety gap, highlighting the crucial need for human oversight and the implementation of AI safety guardrails.

Item Response Theory (IRT) has been widely used in educational psychometrics to assess student ability, as well as the difficulty and discrimination of test questions. In this context, discrimination specifically refers to how effectively a question distinguishes between students of different ability levels, and it does not carry any connotation related to fairness. In recent years, IRT has been successfully used to evaluate the predictive performance of Machine Learning (ML) models, but this paper marks its first application in fairness evaluation. In this paper, we propose a novel Fair-IRT framework to evaluate a set of predictive models on a set of individuals, while simultaneously eliciting specific parameters, namely, the ability to make fair predictions (a feature of predictive models), as well as the discrimination and difficulty of individuals that affect the prediction results. Furthermore, we conduct a series of experiments to comprehensively understand the implications of these parameters for fairness evaluation. Detailed explanations for item characteristic curves (ICCs) are provided for particular individuals. We propose the flatness of ICCs to disentangle the unfairness between individuals and predictive models. The experiments demonstrate the effectiveness of this framework as a fairness evaluation tool. Two real-world case studies illustrate its potential application in evaluating fairness in both classification and regression tasks. Our paper aligns well with the Responsible Web track by proposing a Fair-IRT framework to evaluate fairness in ML models, which directly contributes to the development of a more inclusive, equitable, and trustworthy AI.

Objective: To improve prediction of Chronic Kidney Disease (CKD) progression to End Stage Renal Disease (ESRD) using machine learning (ML) and deep learning (DL) models applied to an integrated clinical and claims dataset of varying observation windows, supported by explainable AI (XAI) to enhance interpretability and reduce bias. Materials and Methods: We utilized data about 10,326 CKD patients, combining their clinical and claims information from 2009 to 2018. Following data preprocessing, cohort identification, and feature engineering, we evaluated multiple statistical, ML and DL models using data extracted from five distinct observation windows. Feature importance and Shapley value analysis were employed to understand key predictors. Models were tested for robustness, clinical relevance, misclassification errors and bias issues. Results: Integrated data models outperformed those using single data sources, with the Long Short-Term Memory (LSTM) model achieving the highest AUC (0.93) and F1 score (0.65). A 24-month observation window was identified as optimal for balancing early detection and prediction accuracy. The 2021 eGFR equation improved prediction accuracy and reduced racial bias, notably for African American patients. Discussion: Improved ESRD prediction accuracy, results interpretability and bias mitigation strategies presented in this study have the potential to significantly enhance CKD and ESRD management, support targeted early interventions and reduce healthcare disparities. Conclusion: This study presents a robust framework for predicting ESRD outcomes in CKD patients, improving clinical decision-making and patient care through multi-sourced, integrated data and AI/ML methods. Future research will expand data integration and explore the application of this framework to other chronic diseases.

PoliTune: Analyzing the Impact of Data Selection and Fine-Tuning on Economic and Political Biases in Large Language Models

Ahmed Agiza, Mohamed Mostagir, Sherief Reda

In an era where language models are increasingly integrated into decision-making and communication, understanding the biases within Large Language Models (LLMs) becomes imperative, especially when these models are applied in the economic and political domains. This work investigates the impact of fine-tuning and data selection on economic and political biases in LLMs. In this context, we introduce PoliTune, a fine-tuning methodology to explore the systematic aspects of aligning LLMs with specific ideologies, mindful of the biases that arise from their extensive training on diverse datasets. Distinct from earlier efforts that either focus on smaller models or entail resource-intensive pre-training, PoliTune employs Parameter-Efficient Fine-Tuning (PEFT) techniques, which allow for the alignment of LLMs with targeted ideologies by modifying a small subset of parameters. We introduce a systematic method for using the open-source LLM Llama3-70B for dataset selection, annotation, and synthesizing a preferences dataset for Direct Preference Optimization (DPO) to align the model with a given political ideology. We assess the effectiveness of PoliTune through both quantitative and qualitative evaluations of aligning open-source LLMs (Llama3-8B and Mistral-7B) to different ideologies. Our work analyzes the potential of embedding specific biases into LLMs and contributes to the dialogue on the ethical application of AI, highlighting the importance of deploying AI in a manner that aligns with societal values.

Interpretable Machine Learning for Survival Analysis

Sophie Hanna Langbein, Mateusz Krzyziński, Mikołaj Spytek +3

With the spread and rapid advancement of black box machine learning models, the field of interpretable machine learning (IML) or explainable artificial intelligence (XAI) has become increasingly important over the last decade. This is particularly relevant for survival analysis, where the adoption of IML techniques promotes transparency, accountability and fairness in sensitive areas, such as clinical decision making processes, the development of targeted therapies, interventions or in other medical or healthcare related contexts. More specifically, explainability can uncover a survival model's potential biases and limitations and provide more mathematically sound ways to understand how and which features are influential for prediction or constitute risk factors. However, the lack of readily available IML methods may have deterred medical practitioners and policy makers in public health from leveraging the full potential of machine learning for predicting time-to-event data. We present a comprehensive review of the limited existing amount of work on IML methods for survival analysis within the context of the general IML taxonomy. In addition, we formally detail how commonly used IML methods, such as such as individual conditional expectation (ICE), partial dependence plots (PDP), accumulated local effects (ALE), different feature importance measures or Friedman's H-interaction statistics can be adapted to survival outcomes. An application of several IML methods to real data on data on under-5 year mortality of Ghanaian children from the Demographic and Health Surveys (DHS) Program serves as a tutorial or guide for researchers, on how to utilize the techniques in practice to facilitate understanding of model decisions or predictions.

The Role of Subgroup Separability in Group-Fair Medical Image Classification

Charles Jones, Mélanie Roschewitz, Ben Glocker

We investigate performance disparities in deep classifiers. We find that the ability of classifiers to separate individuals into subgroups varies substantially across medical imaging modalities and protected characteristics; crucially, we show that this property is predictive of algorithmic bias. Through theoretical analysis and extensive empirical evaluation, we find a relationship between subgroup separability, subgroup disparities, and performance degradation when models are trained on data with systematic bias such as underdiagnosis. Our findings shed new light on the question of how models become biased, providing important insights for the development of fair medical imaging AI.

As debates on potential societal harm from artificial intelligence (AI) culminate in legislation and international norms, a global divide is emerging in both AI regulatory frameworks and international governance structures. In terms of local regulatory frameworks, the European Union (E.U.), Canada, and Brazil follow a horizontal or lateral approach that postulates the homogeneity of AI, seeks to identify common causes of harm, and demands uniform human interventions. In contrast, the United States (U.S.), the United Kingdom (U.K.), Israel, and Switzerland (and potentially China) have pursued a context-specific or modular approach, tailoring regulations to the specific use cases of AI systems. This paper argues for a context-specific approach to effectively address evolving risks in diverse mission-critical domains, while avoiding social costs associated with one-size-fits-all approaches. However, to enhance the systematicity and interoperability of international norms and accelerate global harmonization, this paper proposes an alternative contextual, coherent, and commensurable (3C) framework. To ensure contextuality, the framework (i) bifurcates the AI life cycle into two phases: learning and deployment for specific tasks, instead of defining foundation or general-purpose models; and (ii) categorizes these tasks based on their application and interaction with humans as follows: autonomous, discriminative (allocative, punitive, and cognitive), and generative AI. To ensure coherency, each category is assigned specific regulatory objectives replacing 2010s vintage AI ethics. To ensure commensurability, the framework promotes the adoption of international standards for measuring and mitigating risks.

We investigate opinion dynamics and information spreading on networks under the influence of content filtering technologies. The filtering mechanism, present in many online social platforms, reduces individuals' exposure to disagreeing opinions, producing algorithmic bias. We derive evolution equations for global opinion variables in the presence of algorithmic bias, network community structure, noise (independent behavior of individuals), and pairwise or group interactions. We consider the case where the social platform shows a predilection for one opinion over its opposite, unbalancing the dynamics in favor of that opinion. We show that if the imbalance is strong enough, it may determine the final global opinion and the dynamical behavior of the population. We find a complex phase diagram including phases of coexistence, consensus, and polarization of opinions as possible final states of the model, with phase transitions of different order between them. The fixed point structure of the equations determines the dynamics to a large extent. We focus on the time needed for convergence and conclude that this quantity varies within a wide range, showing occasionally signatures of critical slowing down and meta-stability.

We consider the task of enforcing individual fairness in gradient boosting. Gradient boosting is a popular method for machine learning from tabular data, which arise often in applications where algorithmic fairness is a concern. At a high level, our approach is a functional gradient descent on a (distributionally) robust loss function that encodes our intuition of algorithmic fairness for the ML task at hand. Unlike prior approaches to individual fairness that only work with smooth ML models, our approach also works with non-smooth models such as decision trees. We show that our algorithm converges globally and generalizes. We also demonstrate the efficacy of our algorithm on three ML problems susceptible to algorithmic bias.

Measuring Social Biases of Crowd Workers using Counterfactual Queries

Bhavya Ghai, Q. Vera Liao, Yunfeng Zhang +1

Social biases based on gender, race, etc. have been shown to pollute machine learning (ML) pipeline predominantly via biased training datasets. Crowdsourcing, a popular cost-effective measure to gather labeled training datasets, is not immune to the inherent social biases of crowd workers. To ensure such social biases aren't passed onto the curated datasets, it's important to know how biased each crowd worker is. In this work, we propose a new method based on counterfactual fairness to quantify the degree of inherent social bias in each crowd worker. This extra information can be leveraged together with individual worker responses to curate a less biased dataset.

As artificial intelligence (AI) tools become widely adopted, large language models (LLMs) are increasingly involved on both sides of decision-making processes, ranging from hiring to content moderation. This dual adoption raises a critical question: do LLMs systematically favor content that resembles their own outputs? Prior research in computer science has identified self-preference bias -- the tendency of LLMs to favor their own generated content -- but its real-world implications have not been empirically evaluated. We focus on the hiring context, where job applicants often rely on LLMs to refine resumes, while employers deploy them to screen those same resumes. Using a large-scale controlled resume correspondence experiment, we find that LLMs consistently prefer resumes generated by themselves over those written by humans or produced by alternative models, even when content quality is controlled. The bias against human-written resumes is particularly substantial, with self-preference bias ranging from 67% to 82% across major commercial and open-source models. To assess labor market impact, we simulate realistic hiring pipelines across 24 occupations. These simulations show that candidates using the same LLM as the evaluator are 23% to 60% more likely to be shortlisted than equally qualified applicants submitting human-written resumes, with the largest disadvantages observed in business-related fields such as sales and accounting. We further demonstrate that this bias can be reduced by more than 50% through simple interventions targeting LLMs' self-recognition capabilities. These findings highlight an emerging but previously overlooked risk in AI-assisted decision making and call for expanded frameworks of AI fairness that address not only demographic-based disparities, but also biases in AI-AI interactions.