Nil-Jana Akpinar

Nil-Jana Akpinar, Manish Nagireddy, Logan Stapleton et al.

Motivated by the growing importance of reducing unfairness in ML predictions, Fair-ML researchers have presented an extensive suite of algorithmic 'fairness-enhancing' remedies. Most existing algorithms, however, are agnostic to the sources of the observed unfairness. As a result, the literature currently lacks guiding frameworks to specify conditions under which each algorithmic intervention can potentially alleviate the underpinning cause of unfairness. To close this gap, we scrutinize the underlying biases (e.g., in the training data or design choices) that cause observational unfairness. We present the conceptual idea and a first implementation of a bias-injection sandbox tool to investigate fairness consequences of various biases and assess the effectiveness of algorithmic remedies in the presence of specific types of bias. We call this process the bias(stress)-testing of algorithmic interventions. Unlike existing toolkits, ours provides a controlled environment to counterfactually inject biases in the ML pipeline. This stylized setup offers the distinct capability of testing fairness interventions beyond observational data and against an unbiased benchmark. In particular, we can test whether a given remedy can alleviate the injected bias by comparing the predictions resulting after the intervention in the biased setting with true labels in the unbiased regime-that is, before any bias injection. We illustrate the utility of our toolkit via a proof-of-concept case study on synthetic data. Our empirical analysis showcases the type of insights that can be obtained through our simulations.

Vibhhu Sharma, Shantanu Gupta, Nil-Jana Akpinar et al.

As recommender systems become widely deployed in different domains, they increasingly influence their users' beliefs and preferences. Auditing recommender systems is crucial as it not only ensures the continuous improvement of recommendation algorithms but also safeguards against potential issues like biases and ethical concerns. In this paper, we view recommender system auditing from a causal lens and provide a general recipe for defining auditing metrics. Under this general causal auditing framework, we categorize existing auditing metrics and identify gaps in them -- notably, the lack of metrics for auditing user agency while accounting for the multi-step dynamics of the recommendation process. We leverage our framework and propose two classes of such metrics:future- and past-reacheability and stability, that measure the ability of a user to influence their own and other users' recommendations, respectively. We provide both a gradient-based and a black-box approach for computing these metrics, allowing the auditor to compute them under different levels of access to the recommender system. In our experiments, we demonstrate the efficacy of methods for computing the proposed metrics and inspect the design of recommender systems through these proposed metrics.

Valeriia Cherepanova, Chia-Jung Lee, Nil-Jana Akpinar et al.

Large language models (LLMs) have been shown to be effective on tabular prediction tasks in the low-data regime, leveraging their internal knowledge and ability to learn from instructions and examples. However, LLMs can fail to generate predictions that satisfy group fairness, that is, produce equitable outcomes across groups. Critically, conventional debiasing approaches for natural language tasks do not directly translate to mitigating group unfairness in tabular settings. In this work, we systematically investigate four empirical approaches to improve group fairness of LLM predictions on tabular datasets, including fair prompt optimization, soft prompt tuning, strategic selection of few-shot examples, and self-refining predictions via chain-of-thought reasoning. Through experiments on four tabular datasets using both open-source and proprietary LLMs, we show the effectiveness of these methods in enhancing demographic parity while maintaining high overall performance. Our analysis provides actionable insights for practitioners in selecting the most suitable approach based on their specific requirements and constraints.

Nil-Jana Akpinar, Sandeep Avula, CJ Lee et al.

Large Language Models (LLMs) are increasingly used to generate and edit scientific abstracts, yet their integration into academic writing raises questions about trust, quality, and disclosure. Despite growing adoption, little is known about how readers perceive LLM-generated summaries and how these perceptions influence evaluations of scientific work. This paper presents a mixed-methods survey experiment investigating whether readers with ML expertise can distinguish between human- and LLM-generated abstracts, how actual and perceived LLM involvement affects judgments of quality and trustworthiness, and what orientations readers adopt toward AI-assisted writing. Our findings show that participants struggle to reliably identify LLM-generated content, yet their beliefs about LLM involvement significantly shape their evaluations. Notably, abstracts edited by LLMs are rated more favorably than those written solely by humans or LLMs. We also identify three distinct reader orientations toward LLM-assisted writing, offering insights into evolving norms and informing policy around disclosure and acceptable use in scientific communication.

Vivian Lai, Zana Buçinca, Nil-Jana Akpinar et al.

Proactive AI writing assistants need to predict when users want drafting help, yet we lack empirical understanding of what drives preferences. Through a factorial vignette study with 50 participants making 750 pairwise comparisons, we find compositional effort dominates decisions ($ρ= 0.597$) while urgency shows no predictive power ($ρ\approx 0$). More critically, users exhibit a striking perception-behavior gap: they rank urgency first in self-reports despite it being the weakest behavioral driver, representing a complete preference inversion. This misalignment has measurable consequences. Systems designed from users' stated preferences achieve only 57.7\% accuracy, underperforming even naive baselines, while systems using behavioral patterns reach significantly higher 61.3\% ($p < 0.05$). These findings demonstrate that relying on user introspection for system design actively misleads optimization, with direct implications for proactive natural language generation (NLG) systems.

Jacqueline Maasch, Kyra Gan, Violet Chen et al. · mit

Identifying the causal pathways of unfairness is a critical objective for improving policy design and algorithmic decision-making. Prior work in causal fairness analysis often requires knowledge of the causal graph, hindering practical applications in complex or low-knowledge domains. Moreover, global discovery methods that learn causal structure from data can display unstable performance on finite samples, preventing robust fairness conclusions. To mitigate these challenges, we introduce local discovery for direct discrimination (LD3): a method that uncovers structural evidence of direct unfairness by identifying the causal parents of an outcome variable. LD3 performs a linear number of conditional independence tests relative to variable set size, and allows for latent confounding under the sufficient condition that all parents of the outcome are observed. We show that LD3 returns a valid adjustment set (VAS) under a new graphical criterion for the weighted controlled direct effect, a qualitative indicator of direct discrimination. LD3 limits unnecessary adjustment, providing interpretable VAS for assessing unfairness. We use LD3 to analyze causal fairness in two complex decision systems: criminal recidivism prediction and liver transplant allocation. LD3 was more time-efficient and returned more plausible results on real-world data than baselines, which took 46$\times$ to 5870$\times$ longer to execute.

Nil-Jana Akpinar, Chia-Jung Lee, Vanessa Murdock et al.

Large Language Models (LLMs) should answer factual questions truthfully, grounded in objective knowledge, regardless of user context such as self-disclosed personal information, or system personalization. In this paper, we present the first systematic evaluation of LLM robustness to inquiry personas, i.e. user profiles that convey attributes like identity, expertise, or belief. While prior work has primarily focused on adversarial inputs or distractors for robustness testing, we evaluate plausible, human-centered inquiry persona cues that users disclose in real-world interactions. We find that such cues can meaningfully alter QA accuracy and trigger failure modes such as refusals, hallucinated limitations, and role confusion. These effects highlight how model sensitivity to user framing can compromise factual reliability, and position inquiry persona testing as an effective tool for robustness evaluation.

Nil-Jana Akpinar, Zachary C. Lipton, Alexandra Chouldechova

Predictive risk models in the public sector are commonly developed using administrative data that is more complete for subpopulations that more greatly rely on public services. In the United States, for instance, information on health care utilization is routinely available to government agencies for individuals supported by Medicaid and Medicare, but not for the privately insured. Critiques of public sector algorithms have identified such differential feature under-reporting as a driver of disparities in algorithmic decision-making. Yet this form of data bias remains understudied from a technical viewpoint. While prior work has examined the fairness impacts of additive feature noise and features that are clearly marked as missing, the setting of data missingness absent indicators (i.e. differential feature under-reporting) has been lacking in research attention. In this work, we present an analytically tractable model of differential feature under-reporting which we then use to characterize the impact of this kind of data bias on algorithmic fairness. We demonstrate how standard missing data methods typically fail to mitigate bias in this setting, and propose a new set of methods specifically tailored to differential feature under-reporting. Our results show that, in real world data settings, under-reporting typically leads to increasing disparities. The proposed solution methods show success in mitigating increases in unfairness.

Nil-Jana Akpinar, Maria De-Arteaga, Alexandra Chouldechova

Police departments around the world have been experimenting with forms of place-based data-driven proactive policing for over two decades. Modern incarnations of such systems are commonly known as hot spot predictive policing. These systems predict where future crime is likely to concentrate such that police can allocate patrols to these areas and deter crime before it occurs. Previous research on fairness in predictive policing has concentrated on the feedback loops which occur when models are trained on discovered crime data, but has limited implications for models trained on victim crime reporting data. We demonstrate how differential victim crime reporting rates across geographical areas can lead to outcome disparities in common crime hot spot prediction models. Our analysis is based on a simulation patterned after district-level victimization and crime reporting survey data for Bogotá, Colombia. Our results suggest that differential crime reporting rates can lead to a displacement of predicted hotspots from high crime but low reporting areas to high or medium crime and high reporting areas. This may lead to misallocations both in the form of over-policing and under-policing.

Nil-Jana Akpinar, Aaditya Ramdas, Umut Acar

Educational software data promises unique insights into students' study behaviors and drivers of success. While much work has been dedicated to performance prediction in massive open online courses, it is unclear if the same methods can be applied to blended courses and a deeper understanding of student strategies is often missing. We use pattern mining and models borrowed from Natural Language Processing (NLP) to understand student interactions and extract frequent strategies from a blended college course. Fine-grained clickstream data is collected through Diderot, a non-commercial educational support system that spans a wide range of functionalities. We find that interaction patterns differ considerably based on the assessment type students are preparing for, and many of the extracted features can be used for reliable performance prediction. Our results suggest that the proposed hybrid NLP methods can provide valuable insights even in the low-data setting of blended courses given enough data granularity.

Nil-Jana Akpinar, Bernhard Kratzwald, Stefan Feuerriegel

Learning to predict solutions to real-valued combinatorial graph problems promises efficient approximations. As demonstrated based on the NP-hard edge clique cover number, recurrent neural networks (RNNs) are particularly suited for this task and can even outperform state-of-the-art heuristics. However, the theoretical framework for estimating real-valued RNNs is understood only poorly. As our primary contribution, this is the first work that upper bounds the sample complexity for learning real-valued RNNs. While such derivations have been made earlier for feed-forward and convolutional neural networks, our work presents the first such attempt for recurrent neural networks. Given a single-layer RNN with $a$ rectified linear units and input of length $b$, we show that a population prediction error of $\varepsilon$ can be realized with at most $\tilde{\mathcal{O}}(a^4b/\varepsilon^2)$ samples. We further derive comparable results for multi-layer RNNs. Accordingly, a size-adaptive RNN fed with graphs of at most $n$ vertices can be learned in $\tilde{\mathcal{O}}(n^6/\varepsilon^2)$, i.e., with only a polynomial number of samples. For combinatorial graph problems, this provides a theoretical foundation that renders RNNs competitive.