There is a growing demand for transparency in search engines to understand how search results are curated and to enhance users’ trust. Prior research has introduced search result explanations with a focus on how to explain, assuming explanations are beneficial. Our study takes a step back to examine if search explanations are needed and when they are likely to provide benefits. Additionally, we summarize key characteristics of helpful explanations and share users’ perspectives on explanation features provided by Google and Bing. Interviews with non-technical individuals reveal that users do not always seek or understand search explanations and mostly desire them for complex and critical tasks. They find Google’s search explanations too obvious but appreciate the ability to contest search results. Based on our findings, we offer design recommendations for search engines and explanations to help users better evaluate search results and enhance their search experience.

Ruijia Cheng, Alison Smith-Renner, Ke Zhang, Joel Tetreault, Alejandro Jaimes-Larrarte. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Vivian Lai, Alison Smith-Renner, Ke Zhang, Ruijia Cheng, Wenjuan Zhang, Joel Tetreault, Alejandro Jaimes-Larrarte. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Explainable AI (XAI) seeks to disclose how an AI system arrives at its outcomes. But the nature of the disclosure depends in part on who needs to understand the AI and the available explanation modalities (e.g., verbal and visual). Users’ preferences regarding explanation modalities might differ, as some might prefer spoken explanations compared to visual ones. However, we argue for broadening the explanation modalities, to consider also tangible and physical forms. In traditional product design, physical forms have mediated people’s interactions with objects; more recently interacting with physical forms has become prominent with IoT and smart devices, such as smart lighting and robotic vacuum cleaners. But how tangible interaction can support AI explanations is not yet well understood.

Since the advent of Artificial Intelligence (AI) and Machine Learning (ML), researchers have asked how intelligent computing systems could interact with and relate to their users and their surroundings, leading to debates around issues of biased AI systems, ML black-box, user trust, user’s perception of control over the system, and system’s transparency, to name a few. All of these issues are related to how humans interact with AI or ML systems, through an interface which uses different interaction modalities. Prior studies address these issues from a variety of perspectives, spanning from understanding and framing the problems through ethics and Science and Technology Studies (STS) perspectives to finding effective technical solutions to the problems. But what is shared among almost all those efforts is an assumption that if systems can explain the how and why of their predictions, people will have a better perception of control and therefore will trust such systems more, and even can correct their shortcomings. This research field has been called Explainable AI (XAI). In this studio, we take stock on prior efforts in this area; however, we focus on using Tangible and Embodied Interaction (TEI) as an interaction modality for understanding ML. We note that the affordances of physical forms and their behaviors potentially can not only contribute to the explainability of ML systems, but also can contribute to an open environment for criticism. This studio seeks to both critique explainable ML terminology and to map the opportunities that TEI can offer to the HCI for designing more sustainable, graspable and just intelligent systems.

Automatically generated explanations of how machine learning (ML) models reason can help users understand and accept them. However, explanations can have unintended consequences: promoting over-reliance or undermining trust. This paper investigates how explanations shape users' perceptions of ML models with or without the ability to provide feedback to them: (1) does revealing model flaws increase users' desire to "fix" them; (2) does providing explanations cause users to believe - wrongly - that models are introspective, and will thus improve over time. Through two controlled experiments - varying model quality - we show how the combination of explanations and user feedback impacted perceptions, such as frustration and expectations of model improvement. Explanations without opportunity for feedback were frustrating with a lower quality model, while interactions between explanation and feedback for the higher quality model suggest that detailed feedback should not be requested without explanation. Users expected model correction, regardless of whether they provided feedback or received explanations.

We explore predictability and control in interactive systems where controls are easy to validate. Human-in-the-loop techniques allow users to guide unsupervised algorithms by exposing and supporting interaction with underlying model representations, increasing transparency and promising fine-grained control. However, these models must balance user input and the underlying data, meaning they sometimes update slowly, poorly, or unpredictably---either by not incorporating user input as expected (adherence) or by making other unexpected changes (instability). While prior work exposes model internals and supports user feedback, less attention has been paid to users' reactions when transparent models limit control. Focusing on interactive topic models, we explore user perceptions of control using a study where 100 participants organize documents with one of three distinct topic modeling approaches. These approaches incorporate input differently, resulting in varied adherence, stability, update speeds, and model quality. Participants disliked slow updates most, followed by lack of adherence. Instability was polarizing: some participants liked it when it surfaced interesting information, while others did not. Across modeling approaches, participants differed only in whether they noticed adherence.

To address the lack of comparative evaluation of Human-in-the-Loop Topic Modeling (HLTM) systems, we implement and evaluate three contrasting HLTM modeling approaches using simulation experiments.These approaches extend previously proposed frameworks, including constraints and informed prior-based methods.Users should have a sense of control in HLTM systems, so we propose a control metric to measure whether refinement operations' results match users' expectations.Informed prior-based methods provide better control than constraints, but constraints yield higher quality topics.

Human-in-the-loop topic modeling allows users to guide the creation of topic models and to improve model quality without having to be experts in topic modeling algorithms. Prior work in this area has focused either on algorithmic implementation without understanding how users actually wish to improve the model or on user needs but without the context of a fully interactive system. To address this disconnect, we implemented a set of model refinements requested by users in prior work and conducted a study with twelve non-expert participants to examine how end users are affected by issues that arise with a fully interactive, user-centered system. As these issues mirror those identified in interactive machine learning more broadly, such as unpredictability, latency, and trust, we also examined interactive machine learning challenges with non-expert end users through the lens of human-in-the-loop topic modeling. We found that although users experience unpredictability, their reactions vary from positive to negative, and, surprisingly, we did not find any cases of distrust, but instead noted instances where users perhaps trusted the system too much or had too little confidence in themselves.

Topic modeling is a common tool for understanding large bodies of text, but is typically provided as a “take it or leave it” proposition. Incorporating human knowledge in unsupervised learning is a promising approach to create high-quality topic models. Existing interactive systems and modeling algorithms support a wide range of refinement operations to express feedback. However, these systems’ interactions are primarily driven by algorithmic convenience, ignoring users who may lack expertise in topic modeling. To better understand how non-expert users understand, assess, and refine topics, we conducted two user studies—an in-person interview study and an online crowdsourced study. These studies demonstrate a disconnect between what non-expert users want and the complex, low-level operations that current interactive systems support . In particular, our findings include: (1) analysis of how non-expert users perceive topic models; (2) characterization of primary refinement operations expected by non-expert users and ordered by relative preference; (3) further evidence of the benefits of supporting users in directly refining a topic model; (4) design implications for future human-in-the-loop topic modeling interfaces.

Probabilistic topic models are important tools for indexing, summarizing, and analyzing large document collections by their themes. However, promoting end-user understanding of topics remains an open research problem. We compare labels generated by users given four topic visualization techniques—word lists, word lists with bars, word clouds, and network graphs—against each other and against automatically generated labels. Our basis of comparison is participant ratings of how well labels describe documents from the topic. Our study has two phases: a labeling phase where participants label visualized topics and a validation phase where different participants select which labels best describe the topics’ documents. Although all visualizations produce similar quality labels, simple visualizations such as word lists allow participants to quickly understand topics, while complex visualizations take longer but expose multi-word expressions that simpler visualizations obscure. Automatic labels lag behind user-created labels, but our dataset of manually labeled topics highlights linguistic patterns (e.g., hypernyms, phrases) that can be used to improve automatic topic labeling algorithms.

Social media, particularly Twitter, provides an abundance of real-time data. To account for this volume, researchers often use automated analysis and visualization techniques to produce a high-level overview of a Twitter stream. Existing techniques for understanding Twitter data make use of hashtags or word-pairs and may ignore the complex trends in discussions over time. To remedy this, we present an application of statistical topic modeling and alignment (binned topic models) to group related tweets into automatically generated topics and TopicFlow, an interactive tool to visualize the evolution of these topics. The effectiveness of this visualization for reasoning about large data sets is demonstrated by a usability study with 18 participants.