Adam Perer

Anna Kawakami, Venkatesh Sivaraman, Hao-Fei Cheng et al. · cmu

AI-based decision support tools (ADS) are increasingly used to augment human decision-making in high-stakes, social contexts. As public sector agencies begin to adopt ADS, it is critical that we understand workers' experiences with these systems in practice. In this paper, we present findings from a series of interviews and contextual inquiries at a child welfare agency, to understand how they currently make AI-assisted child maltreatment screening decisions. Overall, we observe how workers' reliance upon the ADS is guided by (1) their knowledge of rich, contextual information beyond what the AI model captures, (2) their beliefs about the ADS's capabilities and limitations relative to their own, (3) organizational pressures and incentives around the use of the ADS, and (4) awareness of misalignments between algorithmic predictions and their own decision-making objectives. Drawing upon these findings, we discuss design implications towards supporting more effective human-AI decision-making.

Ángel Alexander Cabrera, Adam Perer, Jason I. Hong · cmu

People work with AI systems to improve their decision making, but often under- or over-rely on AI predictions and perform worse than they would have unassisted. To help people appropriately rely on AI aids, we propose showing them behavior descriptions, details of how AI systems perform on subgroups of instances. We tested the efficacy of behavior descriptions through user studies with 225 participants in three distinct domains: fake review detection, satellite image classification, and bird classification. We found that behavior descriptions can increase human-AI accuracy through two mechanisms: helping people identify AI failures and increasing people's reliance on the AI when it is more accurate. These findings highlight the importance of people's mental models in human-AI collaboration and show that informing people of high-level AI behaviors can significantly improve AI-assisted decision making.

Hong Shen, Ángel Alexander Cabrera, Adam Perer et al. · cmu

This position paper offers a framework to think about how to better involve human influence in algorithmic decision-making of contentious public policy issues. Drawing from insights in communication literature, we introduce a "public(s)-in-the-loop" approach and enumerates three features that are central to this approach: publics as plural political entities, collective decision-making through deliberation, and the construction of publics. It explores how these features might advance our understanding of stakeholder participation in AI design in contentious public policy domains such as recidivism prediction. Finally, it sketches out part of a research agenda for the HCI community to support this work.

Katelyn Morrison, Philipp Spitzer, Violet Turri et al.

Explainability techniques are rapidly being developed to improve human-AI decision-making across various cooperative work settings. Consequently, previous research has evaluated how decision-makers collaborate with imperfect AI by investigating appropriate reliance and task performance with the aim of designing more human-centered computer-supported collaborative tools. Several human-centered explainable AI (XAI) techniques have been proposed in hopes of improving decision-makers' collaboration with AI; however, these techniques are grounded in findings from previous studies that primarily focus on the impact of incorrect AI advice. Few studies acknowledge the possibility of the explanations being incorrect even if the AI advice is correct. Thus, it is crucial to understand how imperfect XAI affects human-AI decision-making. In this work, we contribute a robust, mixed-methods user study with 136 participants to evaluate how incorrect explanations influence humans' decision-making behavior in a bird species identification task, taking into account their level of expertise and an explanation's level of assertiveness. Our findings reveal the influence of imperfect XAI and humans' level of expertise on their reliance on AI and human-AI team performance. We also discuss how explanations can deceive decision-makers during human-AI collaboration. Hence, we shed light on the impacts of imperfect XAI in the field of computer-supported cooperative work and provide guidelines for designers of human-AI collaboration systems.

Yanwu Xu, Li Sun, Wei Peng et al.

This paper introduces an innovative methodology for producing high-quality 3D lung CT images guided by textual information. While diffusion-based generative models are increasingly used in medical imaging, current state-of-the-art approaches are limited to low-resolution outputs and underutilize radiology reports' abundant information. The radiology reports can enhance the generation process by providing additional guidance and offering fine-grained control over the synthesis of images. Nevertheless, expanding text-guided generation to high-resolution 3D images poses significant memory and anatomical detail-preserving challenges. Addressing the memory issue, we introduce a hierarchical scheme that uses a modified UNet architecture. We start by synthesizing low-resolution images conditioned on the text, serving as a foundation for subsequent generators for complete volumetric data. To ensure the anatomical plausibility of the generated samples, we provide further guidance by generating vascular, airway, and lobular segmentation masks in conjunction with the CT images. The model demonstrates the capability to use textual input and segmentation tasks to generate synthesized images. The results of comparative assessments indicate that our approach exhibits superior performance compared to the most advanced models based on GAN and diffusion techniques, especially in accurately retaining crucial anatomical features such as fissure lines, airways, and vascular structures. This innovation introduces novel possibilities. This study focuses on two main objectives: (1) the development of a method for creating images based on textual prompts and anatomical components, and (2) the capability to generate new images conditioning on anatomical elements. The advancements in image generation can be applied to enhance numerous downstream tasks.

Yuzhe Lu, Adam Perer

Deep learning methods, in particular convolutional neural networks, have emerged as a powerful tool in medical image computing tasks. While these complex models provide excellent performance, their black-box nature may hinder real-world adoption in high-stakes decision-making. In this paper, we propose an interactive system to take advantage of state-of-the-art interpretability techniques to assist radiologists with breast cancer screening. Our system integrates a deep learning model into the radiologists' workflow and provides novel interactions to promote understanding of the model's decision-making process. Moreover, we demonstrate that our system can take advantage of user interactions progressively to provide finer-grained explainability reports with little labeling overhead. Due to the generic nature of the adopted interpretability technique, our system is domain-agnostic and can be used for many different medical image computing tasks, presenting a novel perspective on how we can leverage visual analytics to transform originally static interpretability techniques to augment human decision making and promote the adoption of medical AI.

Venkatesh Sivaraman, Eric P. Mason, Mengfan Ellen Li et al.

Artificial intelligence (AI)-based decision support systems can be highly accurate yet still fail to support users or improve decisions. Existing theories of AI-assisted decision-making focus on calibrating reliance on AI advice, leaving it unclear how different system designs might influence the reasoning processes underneath. We address this gap by reconsidering AI interfaces as collections of intelligent reasoning cues: discrete pieces of AI information that can individually influence decision-making. We then explore the roles of eight types of reasoning cues in a high-stakes clinical decision (treating patients with sepsis in intensive care). Through contextual inquiries with six teams and a think-aloud study with 25 physicians, we find that reasoning cues have distinct patterns of influence that can directly inform design. Our results also suggest that reasoning cues should prioritize tasks with high variability and discretion, adapt to ensure compatibility with evolving decision needs, and provide complementary, rigorous insights on complex cases.

Anthony Hughes, Alexander Goldberg, Prince Jha et al.

Safety classifiers are essential safeguards within generative AI systems, filtering harmful content or identifying at-risk users when interacting with large language models. Despite their necessity, these models are trained on sensitive datasets including discussions of self-harm and mental health, raising important, yet poorly understood, privacy concerns. Membership inference attacks (MIAs) allow adversaries to infer membership of examples used to train models. In this work, we hypothesize that identifying the examples on which the classifier is least confident are informative for an adversary to infer membership. This reflects a localized failure of generalization, where the model relies on memorization to resolve ambiguity in the training set. To investigate this, we introduce a new boundary-targeted selection strategy that identifies low confidence examples that amplify the signal of an examples membership within a training set. Our experimental results show that an adversary can recover 19\% of the conversations a safety classifier flagged as indicating user distress, at a 5\% false-positive rate, on a classifier fine-tuned for detecting a user who may require emotional support. This is $3.5$ times more than attacking using state-of-the-art MIA methods alone. Finally, we characterize the boundary laying examples and show that content-based filtering is ineffective for protection, and existing noise strategies can effectively mitigate susceptibility of these examples.

Eunkyu Park, Wesley Hanwen Deng, Cheyon Jin et al.

Vision-Language Models (VLMs) continue to struggle to make morally salient judgments in multimodal and socially ambiguous contexts. Prior works typically rely on binary or pairwise supervision, which often fail to capture the continuous and pluralistic nature of human moral reasoning. We present MM-SCALE (Multimodal Moral Scale), a large-scale dataset for aligning VLMs with human moral preferences through 5-point scalar ratings and explicit modality grounding. Each image-scenario pair is annotated with moral acceptability scores and grounded reasoning labels by humans using an interface we tailored for data collection, enabling listwise preference optimization over ranked scenario sets. By moving from discrete to scalar supervision, our framework provides richer alignment signals and finer calibration of multimodal moral reasoning. Experiments show that VLMs fine-tuned on MM-SCALE achieve higher ranking fidelity and more stable safety calibration than those trained with binary signals.

Venkatesh Sivaraman, Patrick Vossler, Adam Perer et al.

Generative artificial intelligence (AI) tools can now help people perform complex data science tasks regardless of their expertise. While these tools have great potential to help more people work with data, their end-to-end approach does not support users in evaluating alternative approaches and reformulating problems, both critical to solving open-ended tasks in high-stakes domains. In this paper, we reflect on two AI data science systems designed for the medical setting and how they function as tools for thought. We find that success in these systems was driven by constructing AI workflows around intentionally-designed intermediate artifacts, such as readable query languages, concept definitions, or input-output examples. Despite opaqueness in other parts of the AI process, these intermediates helped users reason about important analytical choices, refine their initial questions, and contribute their unique knowledge. We invite the HCI community to consider when and how intermediate artifacts should be designed to promote effective data science thinking.

Denis Newman-Griffis, Venkatesh Sivaraman, Adam Perer et al.

Embeddings of words and concepts capture syntactic and semantic regularities of language; however, they have seen limited use as tools to study characteristics of different corpora and how they relate to one another. We introduce TextEssence, an interactive system designed to enable comparative analysis of corpora using embeddings. TextEssence includes visual, neighbor-based, and similarity-based modes of embedding analysis in a lightweight, web-based interface. We further propose a new measure of embedding confidence based on nearest neighborhood overlap, to assist in identifying high-quality embeddings for corpus analysis. A case study on COVID-19 scientific literature illustrates the utility of the system. TextEssence is available from https://github.com/drgriffis/text-essence.

Venkatesh Sivaraman, Anika Vaishampayan, Xiaotong Li et al.

Temporal predictive models have the potential to improve decisions in health care, public services, and other domains, yet they often fail to effectively support decision-makers. Prior literature shows that many misalignments between model behavior and decision-makers' expectations stem from issues of model specification, namely how, when, and for whom predictions are made. However, model specifications for predictive tasks are highly technical and difficult for non-data-scientist stakeholders to interpret and critique. To address this challenge we developed Tempo, an interactive system that helps data scientists and domain experts collaboratively iterate on model specifications. Using Tempo's simple yet precise temporal query language, data scientists can quickly prototype specifications with greater transparency about pre-processing choices. Moreover, domain experts can assess performance within data subgroups to validate that models behave as expected. Through three case studies, we demonstrate how Tempo helps multidisciplinary teams quickly prune infeasible specifications and identify more promising directions to explore.

Venkatesh Sivaraman, Katelyn Morrison, Will Epperson et al.

As AI-based clinical decision support (AI-CDS) is introduced in more and more aspects of healthcare services, HCI research plays an increasingly important role in designing for complementarity between AI and clinicians. However, current evaluations of AI-CDS often fail to capture when AI is and is not useful to clinicians. This position paper reflects on our work and influential AI-CDS literature to advocate for moving beyond evaluation metrics like Trust, Reliance, Acceptance, and Performance on the AI's task (what we term the "trap" of human-AI collaboration). Although these metrics can be meaningful in some simple scenarios, we argue that optimizing for them ignores important ways that AI falls short of clinical benefit, as well as ways that clinicians successfully use AI. As the fields of HCI and AI in healthcare develop new ways to design and evaluate CDS tools, we call on the community to prioritize ecologically valid, domain-appropriate study setups that measure the emergent forms of value that AI can bring to healthcare professionals.

Katelyn Morrison, Arpit Mathur, Aidan Bradshaw et al.

As text-to-image generative models rapidly improve, AI researchers are making significant advances in developing domain-specific models capable of generating complex medical imagery from text prompts. Despite this, these technical advancements have overlooked whether and how medical professionals would benefit from and use text-to-image generative AI (GenAI) in practice. By developing domain-specific GenAI without involving stakeholders, we risk the potential of building models that are either not useful or even more harmful than helpful. In this paper, we adopt a human-centered approach to responsible model development by involving stakeholders in evaluating and reflecting on the promises, risks, and challenges of a novel text-to-CT Scan GenAI model. Through exploratory model prompting activities, we uncover the perspectives of medical students, radiology trainees, and radiologists on the role that text-to-CT Scan GenAI can play across medical education, training, and practice. This human-centered approach additionally enabled us to surface technical challenges and domain-specific risks of generating synthetic medical images. We conclude by reflecting on the implications of medical text-to-image GenAI.

Unnseo Park, Venkatesh Sivaraman, Adam Perer

Reinforcement learning (RL) is a promising approach to generate treatment policies for sepsis patients in intensive care. While retrospective evaluation metrics show decreased mortality when these policies are followed, studies with clinicians suggest their recommendations are often spurious. We propose that these shortcomings may be due to lack of diversity in observed actions and outcomes in the training data, and we construct experiments to investigate the feasibility of predicting sepsis disease severity changes due to clinician actions. Preliminary results suggest incorporating action information does not significantly improve model performance, indicating that clinician actions may not be sufficiently variable to yield measurable effects on disease progression. We discuss the implications of these findings for optimizing sepsis treatment.

Venkatesh Sivaraman, Yiwei Wu, Adam Perer

Modern machine learning techniques commonly rely on complex, high-dimensional embedding representations to capture underlying structure in the data and improve performance. In order to characterize model flaws and choose a desirable representation, model builders often need to compare across multiple embedding spaces, a challenging analytical task supported by few existing tools. We first interviewed nine embedding experts in a variety of fields to characterize the diverse challenges they face and techniques they use when analyzing embedding spaces. Informed by these perspectives, we developed a novel system called Emblaze that integrates embedding space comparison within a computational notebook environment. Emblaze uses an animated, interactive scatter plot with a novel Star Trail augmentation to enable visual comparison. It also employs novel neighborhood analysis and clustering procedures to dynamically suggest groups of points with interesting changes between spaces. Through a series of case studies with ML experts, we demonstrate how interactive comparison with Emblaze can help gain new insights into embedding space structure.

Donghoon Shin, Sachin Grover, Kenneth Holstein et al.

Explainable AI (XAI) is a promising means of supporting human-AI collaborations for high-stakes visual detection tasks, such as damage detection tasks from satellite imageries, as fully-automated approaches are unlikely to be perfectly safe and reliable. However, most existing XAI techniques are not informed by the understandings of task-specific needs of humans for explanations. Thus, we took a first step toward understanding what forms of XAI humans require in damage detection tasks. We conducted an online crowdsourced study to understand how people explain their own assessments, when evaluating the severity of building damage based on satellite imagery. Through the study with 60 crowdworkers, we surfaced six major strategies that humans utilize to explain their visual damage assessments. We present implications of our findings for the design of XAI methods for such visual detection contexts, and discuss opportunities for future research.

Ángel Alexander Cabrera, Abraham J. Druck, Jason I. Hong et al.

AI systems can fail to learn important behaviors, leading to real-world issues like safety concerns and biases. Discovering these systematic failures often requires significant developer attention, from hypothesizing potential edge cases to collecting evidence and validating patterns. To scale and streamline this process, we introduce crowdsourced failure reports, end-user descriptions of how or why a model failed, and show how developers can use them to detect AI errors. We also design and implement Deblinder, a visual analytics system for synthesizing failure reports that developers can use to discover and validate systematic failures. In semi-structured interviews and think-aloud studies with 10 AI practitioners, we explore the affordances of the Deblinder system and the applicability of failure reports in real-world settings. Lastly, we show how collecting additional data from the groups identified by developers can improve model performance.

Dylan Cashman, Adam Perer, Remco Chang et al.

Deep learning models require the configuration of many layers and parameters in order to get good results. However, there are currently few systematic guidelines for how to configure a successful model. This means model builders often have to experiment with different configurations by manually programming different architectures (which is tedious and time consuming) or rely on purely automated approaches to generate and train the architectures (which is expensive). In this paper, we present Rapid Exploration of Model Architectures and Parameters, or REMAP, a visual analytics tool that allows a model builder to discover a deep learning model quickly via exploration and rapid experimentation of neural network architectures. In REMAP, the user explores the large and complex parameter space for neural network architectures using a combination of global inspection and local experimentation. Through a visual overview of a set of models, the user identifies interesting clusters of architectures. Based on their findings, the user can run ablation and variation experiments to identify the effects of adding, removing, or replacing layers in a given architecture and generate new models accordingly. They can also handcraft new models using a simple graphical interface. As a result, a model builder can build deep learning models quickly, efficiently, and without manual programming. We inform the design of REMAP through a design study with four deep learning model builders. Through a use case, we demonstrate that REMAP allows users to discover performant neural network architectures efficiently using visual exploration and user-defined semi-automated searches through the model space.

Gregory Plumb, Maruan Al-Shedivat, Angel Alexander Cabrera et al.

Most of the work on interpretable machine learning has focused on designing either inherently interpretable models, which typically trade-off accuracy for interpretability, or post-hoc explanation systems, whose explanation quality can be unpredictable. Our method, ExpO, is a hybridization of these approaches that regularizes a model for explanation quality at training time. Importantly, these regularizers are differentiable, model agnostic, and require no domain knowledge to define. We demonstrate that post-hoc explanations for ExpO-regularized models have better explanation quality, as measured by the common fidelity and stability metrics. We verify that improving these metrics leads to significantly more useful explanations with a user study on a realistic task.

Hendrik Strobelt, Sebastian Gehrmann, Michael Behrisch et al.

Neural Sequence-to-Sequence models have proven to be accurate and robust for many sequence prediction tasks, and have become the standard approach for automatic translation of text. The models work in a five stage blackbox process that involves encoding a source sequence to a vector space and then decoding out to a new target sequence. This process is now standard, but like many deep learning methods remains quite difficult to understand or debug. In this work, we present a visual analysis tool that allows interaction with a trained sequence-to-sequence model through each stage of the translation process. The aim is to identify which patterns have been learned and to detect model errors. We demonstrate the utility of our tool through several real-world large-scale sequence-to-sequence use cases.

Josua Krause, Adam Perer, Enrico Bertini

It is commonly believed that increasing the interpretability of a machine learning model may decrease its predictive power. However, inspecting input-output relationships of those models using visual analytics, while treating them as black-box, can help to understand the reasoning behind outcomes without sacrificing predictive quality. We identify a space of possible solutions and provide two examples of where such techniques have been successfully used in practice.