Meetings play a critical infrastructural role in coordinating work. The recent surge of hybrid and remote meetings in computer-mediated spaces has led to new problems (e.g., more time spent in less engaging meetings) and new opportunities (e.g., automated transcription/captioning and recap support). Advances in dialogue summarization offer the potential for improving post-meeting experiences, but fixed-length summaries often fail to meet diverse needs, such as quick overviews or detailed insights. To address these gaps, we use cognitive science and discourse theories to conceptualize two recap designs: important highlights and a structured, hierarchical minutes view, targeting complementary recap needs. We operationalize these representations into high-fidelity prototypes using dialogue summarization. Finally, we evaluate the representations' effectiveness with seven users in the context of their work meetings at Microsoft. Our results show both recap types are valuable in different contexts, enabling collaboration through discussions and consensus-building. Exploring the meaning of users adding, editing, and deleting from recaps suggests varying alignment for using these actions to improve AI-recap. Our design implications, such as incorporating organizational artifacts (e.g., linking presentations) in recaps and personalizing context, advance the discourse of effective recap designs for organizational work and support past results from cognition studies.

Personalization improves user experience by tailoring interactions relevant to each user’s background and preferences. However, personalization requires information about users that platforms often collect without their awareness or their enthusiastic consent. Here, we study how the transparency of AI inferences on users’ personal data affects their privacy decisions and sentiments when sharing data for personalization. We conducted two experiments where participants (N=877) answered questions about themselves for personalized public arts recommendations. Participants indicated their consent to let the system use their inferred data and explicitly provided data after awareness of inferences. Our results show that participants chose restrictive consent decisions for sensitive and incorrect inferences about them and for their answers that led to such inferences. Our findings expand existing privacy discourse to inferences and inform future directions for shaping existing consent mechanisms in light of increasingly pervasive AI inferences.

One useful application of NLP models is to support people in reading complex text from unfamiliar domains (e.g., scientific articles).Simplifying the entire text makes it understandable but sometimes removes important details.On the contrary, helping adult readers understand difficult concepts in context can enhance their vocabulary and knowledge.In a preliminary human study, we first identify that lack of context and unfamiliarity with difficult concepts is a major reason for adult readers' difficulty with domain-specific text.We then introduce targeted concept simplification, a simplification task for rewriting text to help readers comprehend text containing unfamiliar concepts.We also introduce WIKIDOMAINS 1 , a new dataset of 22k definitions from 13 academic domains paired with a difficult concept within each definition.We benchmark the performance of open-source and commercial LLMs, and a simple dictionary baseline on this task across human judgments of ease of understanding and meaning preservation.Interestingly, our human judges preferred explanations about the difficult concept more than simplification of the concept phrase.Further, no single model achieved superior performance across all quality dimensions, and automated metrics also show low correlations with human evaluations of concept simplification (∼ 0.2), opening up rich avenues for research on personalized human reading comprehension support.* Work done as student researcher at Google DeepMind. 1 https://github.com/google-deepmind/wikidomains

We present an empirical study evaluating the quality of multiple-choice questions (MCQs) generated by Large Language Models (LLMs) from a corpus of video transcripts of course lectures in an online data science degree program. With our database of thousands of generated questions, we conducted both human and automated judging of question quality on a representative sample using a broad set of criteria, including well-established Item Writing Flaw (IWF) categories. We found the number of average IWFs per MCQ ranged from 1.6 (rule-based verification) to 2.18 (LLM-based). Among the most frequently identified MCQ flaws were lack of enough context (17%) or answer choices with at least one implausible distractor (57%). Both human and automated assessment identified implausible distractors as one of the most frequent flaw categories. Results from our human annotation study were generally more positive (51--65% good items) compared to our automated assessment study results, which tended toward greater flaw identification (15--25% good items), depending on evaluation method.

Decision Support Systems (DSS) based on Machine Learning (ML) often aim to assist lay decision-makers, who are not math-savvy, in making high-stakes decisions. However, existing ML-based DSS are not always transparent about the probabilistic nature of ML predictions and how uncertain each prediction is. This lack of transparency could give lay decision-makers a false sense of reliability. Growing calls for AI transparency have led to increasing efforts to quantify and communicate model uncertainty. However, there are still gaps in knowledge regarding how and why the decision-makers utilize ML uncertainty information in their decision process. Here, we conducted a qualitative, think-aloud user study with 17 lay decision-makers who interacted with three different DSS: 1) interactive visualization, 2) DSS based on an ML model that provides predictions without uncertainty information, and 3) the same DSS with uncertainty information. Our qualitative analysis found that communicating uncertainty about ML predictions forced participants to slow down and think analytically about their decisions. This in turn made participants more vigilant, resulting in reduction in over-reliance on ML-based DSS. Our work contributes empirical knowledge on how lay decision-makers perceive, interpret, and make use of uncertainty information when interacting with DSS. Such foundational knowledge informs the design of future ML-based DSS that embrace transparent uncertainty communication.

Continuous integration and deployment (CI/CD) is now a widely adopted development model in practice as it reduces the time from ideas to customers. This adoption has also revived the idea of "shifting left" during software development -- a practice intended to find and prevent defects early in the software delivery process. To assist with that, engineering systems integrate developer bots in the development workflow to improve developer productivity and help them identify issues early in the software delivery process.

Wikipedia articles aim to be definitive sources of encyclopedic content. Yet, only 0.6% of Wikipedia articles have high quality according to its quality scale due to insufficient number of Wikipedia editors and enormous number of articles. Supervised Machine Learning (ML) quality improvement approaches that can automatically identify and fix content issues rely on manual labels of individual Wikipedia sentence quality. However, current labeling approaches are tedious and produce noisy labels. Here, we propose an automated labeling approach that identifies the semantic category (e.g., adding citations, clarifications) of historic Wikipedia edits and uses the modified sentences prior to the edit as examples that require that semantic improvement. Highest-rated article sentences are examples that no longer need semantic improvements. We show that training existing sentence quality classification algorithms on our labels improves their performance compared to training them on existing labels. Our work shows that editing behaviors of Wikipedia editors provide better labels than labels generated by crowdworkers who lack the context to make judgments that the editors would agree with.

Today's software development is distributed and involves continuous changes for new features and yet, their development cycle has to be fast and agile. An important component of enabling this agility is selecting the right reviewers for every code-change - the smallest unit of the development cycle. Modern tool-based code review is proven to be an effective way to achieve appropriate code review of software changes. However, the selection of reviewers in these code review systems is at best manual. As software and teams scale, this poses the challenge of selecting the right reviewers, which in turn determines software quality over time. While previous work has suggested automatic approaches to code reviewer recommendations, it has been limited to retrospective analysis. We not only deploy a reviewer suggestions algorithm - WhoDo - and evaluate its effect but also incorporate load balancing as part of it to address one of its major shortcomings: of recommending experienced developers very frequently. We evaluate the effect of this hybrid recommendation + load balancing system on five repositories within Microsoft. Our results are based around various aspects of a commit and how code review affects that. We attempt to quantitatively answer questions which are supposed to play a vital role in effective code review through our data and substantiate it through qualitative feedback of partner repositories.

Quality control is critical to open production communities like Wikipedia. Wikipedia editors enact border quality control with edits (counter-vandalism) and new article creations (new page patrolling) shortly after they are saved. In this paper, we describe a long-standing set of inefficiencies that have plagued new page patrolling by drawing a contrast to the more efficient, distributed processes for counter-vandalism. Further, to address this issue, we demonstrate an effective automated topic model based on a labeling strategy that leverages a folksonomy developed by subject specific working groups in Wikipedia (WikiProject tags) and a flexible ontology (WikiProjects Directory) to arrive at a hierarchical and uniform label set. We are able to attain very high fitness measures (macro ROC-AUC: 95.2%, macro PR-AUC: 74.5%) and real-time performance using word2vec-based features. Finally, we present a proposal for how incorporating this model into current tools will shift the dynamics of new article review positively.