Souti Chattopadhyay

Sadra Sabouri, Zeinabsadat Saghi, Run Huang et al.

Advances in AI agent capabilities have outpaced users' ability to meaningfully oversee their execution. AI agents can perform sophisticated, multi-step knowledge work autonomously from start to finish, yet this process remains effectively inaccessible during execution, often buried within large volumes of intermediate reasoning and outputs: by the time users receive the output, all underlying decisions have already been made without their involvement. This lack of transparency leaves users unable to examine the agent's assumptions, identify errors before they propagate, or redirect execution when it deviates from their intent. The stakes are particularly high in spreadsheet environments, where process and artifact are inseparable. Each decision the agent makes is recorded directly in cells that belong to and reflect on the user. We introduce Pista, a spreadsheet AI agent that decomposes execution into auditable, controllable actions, providing users with visibility into the agent's decision-making process and the capacity to intervene at each step. A formative study (N = 8) and a within-subjects summative evaluation (N = 16) comparing Pista to a baseline agent demonstrated that active participation in execution influenced not only task outcomes but also users' comprehension of the task, their perception of the agent, and their sense of role within the workflow. Users identified their own intent reflected in the agent's actions, detected errors that post-hoc review would have failed to surface, and reported a sense of co-ownership over the resulting output. These findings indicate that meaningful human oversight of AI agents in knowledge work requires not improved post-hoc review mechanisms, but active participation in decisions as they are made.

Zeinabsadat Saghi, Daria Riabukhina, Olubukola Akinbami et al.

Cognitive fatigue, which transitions from focused attention to inexact responses, can cause catastrophic failures in high-stakes environments, yet current black-box assessment techniques ignore the brain's non-Markovian and time-varying interdependent properties, limiting real-time phase transition detection. We develop a fractional dynamical networks-based machine learning (FDNML) framework using coupled fractional-order differential equations to capture brain signal interdependencies and detect cognitive fatigue transitions in real-time. Multifractal properties of brain activity exhibit distinct generalized fractal dimension signatures across fatigue levels, with Wasserstein distances of 0.10, 0.13, and 0.08 between states 0-1, 1-2, and 0-2, respectively. The framework achieves 93.33% classification accuracy and 95% AUROC, enabling the prevention of performance degradation through early detection of neural state transitions.

Zain Merchant, Abrar Anwar, Emily Wang et al.

Navigating unfamiliar environments presents significant challenges for blind and low-vision (BLV) individuals. In this work, we construct a dataset of images and goals across different scenarios such as searching through kitchens or navigating outdoors. We then investigate how grounded instruction generation methods can provide contextually-relevant navigational guidance to users in these instances. Through a sighted user study, we demonstrate that large pretrained language models can produce correct and useful instructions perceived as beneficial for BLV users. We also conduct a survey and interview with 4 BLV users and observe useful insights on preferences for different instructions based on the scenario.

Zeinabsadat Saghi, Run Huang, Souti Chattopadhyay

Creativity is fundamentally human. As AI takes on more of the generative work that once required human imagination, despite documented limitations in creative ability, a critical question emerges: How does GenAI affect users' creativity? Through a within-subject study followed by retrospective interviews with (N=20) programmers, we investigated the impact of LLMs on participants' process of creative thinking in programming and the creativity of generated solutions. Across two conditions (LLM-assisted vs. unassisted), participants using LLMs had significantly shorter idea-generation periods (p=0.0004), leading to fewer creative moments (p=0.002). Qualitative analysis of participants' interactions and interviews revealed four different human-LLM collaboration modes supporting various problem-solving strategies. However, a comparative analysis of the generated solutions shows that while LLMs can help generate more correct and functional code, their solutions contain roughly the same number of ideas as participant-generated ones. Based on our findings, we discuss design implications and considerations for effectively using LLMs to support user creativity.

Amirmohammad Nazari, Sadra Sabouri, Wang Bill Zhu et al.

Many software development tasks, such as implementing features and fixing bugs, begin with developers posing questions about a codebase. However, answering questions about codebases that span millions of lines of code across thousands of files is non-trivial. Standard tools like grep cannot answer questions requiring semantic or inter-procedural reasoning, and large language models (LLMs) struggle with large codebases due to resource and context constraints. In this paper, we present Merlin, a new system for answering free-form questions that require analytical reasoning about code. Merlin integrates an LLM with CodeQL, a program analysis framework that supports expressive queries over large codebases. We face two principal challenges in the design of such systems: First, program analysis queries are diverse and semantically complex; as a result, even syntactically well-formed queries frequently produce degenerate/empty results. Furthermore, relatively few CodeQL queries are available online, limiting the out-of-the-box effectiveness of LLMs as CodeQL query generators. We address these challenges by developing a RAG-based iterative query-generation approach and a novel self-test technique. Our query debugging technique builds on the idea of assistive queries, which generate concrete witnesses that expose and explain semantic flaws in candidate queries. We evaluate Merlin through both experimental and user studies. Over a set of natural language questions derived from common bug-finding tasks, Merlin discovered not only the majority of software issues reported by other approaches, but also issues that would have otherwise remained undetected. Through a within-subject user study, we found that access to Merlin increased task accuracy by an average of 3.8* and simultaneously reduced the time for programmers to complete all tasks by 31%.

Ali Alfageeh, Sadegh AlMahdi Kazemi Zarkouei, Daye Nam et al.

Background and Context. The increasing integration of large language models (LLMs) in computing education presents an emerging challenge in understanding how students use LLMs and craft prompts to solve computational tasks. Prior research has used both qualitative and quantitative methods to analyze prompting behavior, but these approaches lack scalability or fail to effectively capture the semantic evolution of prompts. Objective. In this paper, we investigate whether students prompts can be systematically analyzed using propositional logic constraints. We examine whether this approach can identify patterns in prompt evolution, detect struggling students, and provide insights into effective and ineffective strategies. Method. We introduce Prompt2Constraints, a novel method that translates students prompts into logical constraints. The constraints are able to represent the intent of the prompts in succinct and quantifiable ways. We used this approach to analyze a dataset of 1,872 prompts from 203 students solving introductory programming tasks. Findings. We find that while successful and unsuccessful attempts tend to use a similar number of constraints overall, when students fail, they often modify their prompts more significantly, shifting problem-solving strategies midway. We also identify points where specific interventions could be most helpful to students for refining their prompts. Implications. This work offers a new and scalable way to detect students who struggle in solving natural language programming tasks. This work could be extended to investigate more complex tasks and integrated into programming tools to provide real-time support.

Brihi Joshi, Keyu He, Sahana Ramnath et al.

Language models today are widely used in education, yet their ability to tailor responses for learners with varied informational needs and knowledge backgrounds remains under-explored. To this end, we introduce ELI-Why, a benchmark of 13.4K "Why" questions to evaluate the pedagogical capabilities of language models. We then conduct two extensive human studies to assess the utility of language model-generated explanatory answers (explanations) on our benchmark, tailored to three distinct educational grades: elementary, high-school and graduate school. In our first study, human raters assume the role of an "educator" to assess model explanations' fit to different educational grades. We find that GPT-4-generated explanations match their intended educational background only 50% of the time, compared to 79% for lay human-curated explanations. In our second study, human raters assume the role of a learner to assess if an explanation fits their own informational needs. Across all educational backgrounds, users deemed GPT-4-generated explanations 20% less suited on average to their informational needs, when compared to explanations curated by lay people. Additionally, automated evaluation metrics reveal that explanations generated across different language model families for different informational needs remain indistinguishable in their grade-level, limiting their pedagogical effectiveness.

Philipp Eibl, Sadra Sabouri, Souti Chattopadhyay

Recent AI code assistants have significantly improved their ability to process more complex contexts and generate entire codebases based on a textual description, compared to the popular snippet-level generation. These codebase AI assistants (CBAs) can also extend or adapt codebases, allowing users to focus on higher-level design and deployment decisions. While prior work has extensively studied the impact of snippet-level code generation, this new class of codebase generation models is relatively unexplored. Despite initial anecdotal reports of excitement about these agents, they remain less frequently adopted compared to snippet-level code assistants. To utilize CBAs better, we need to understand how developers interact with CBAs, and how and why CBAs fall short of developers' needs. In this paper, we explored these gaps through a counterbalanced user study and interview with (n = 16) students and developers working on coding tasks with CBAs. We found that participants varied the information in their prompts, like problem description (48% of prompts), required functionality (98% of prompts), code structure (48% of prompts), and their prompt writing process. Despite various strategies, the overall satisfaction score with generated codebases remained low (mean = 2.8, median = 3, on a scale of one to five). Participants mentioned functionality as the most common factor for dissatisfaction (77% of instances), alongside poor code quality (42% of instances) and communication issues (25% of instances). We delve deeper into participants' dissatisfaction to identify six underlying challenges that participants faced when using CBAs, and extracted five barriers to incorporating CBAs into their workflows. Finally, we surveyed 21 commercial CBAs to compare their capabilities with participant challenges and present design opportunities for more efficient and useful CBAs.

Souti Chattopadhyay, Denae Ford, Thomas Zimmermann

Developers are more than "nerds behind computers all day", they lead a normal life, and not all take the traditional path to learn programming. However, the public still sees software development as a profession for "math wizards". To learn more about this special type of knowledge worker from their first-person perspective, we conducted three studies to learn how developers describe a day in their life through vlogs on YouTube and how these vlogs were received by the broader community. We first interviewed 16 developers who vlogged to identify their motivations for creating this content and their intention behind what they chose to portray. Second, we analyzed 130 vlogs (video blogs) to understand the range of the content conveyed through videos. Third, we analyzed 1176 comments from the 130 vlogs to understand the impact the vlogs have on the audience. We found that developers were motivated to promote and build a diverse community, by sharing different aspects of life that define their identity, and by creating awareness about learning and career opportunities in computing. They used vlogs to share a variety of how software developers work and live -- showcasing often unseen experiences, including intimate moments from their personal life. From our comment analysis, we found that the vlogs were valuable to the audience to find information and seek advice. Commenters sought opportunities to connect with others over shared triumphs and trials they faced that were also shown in the vlogs. As a central theme, we found that developers use vlogs to challenge the misconceptions and stereotypes around their identity, work-life, and well-being. These social stigmas are obstacles to an inclusive and accepting community and can deter people from choosing software development as a career. We also discuss the implications of using vlogs to support developers, researchers, and beyond.

Souti Chattopadhyay, Thomas Zimmermann, Denae Ford

Software developers are turning to vlogs (video blogs) to share what a day is like to walk in their shoes. Through these vlogs developers share a rich perspective of their technical work as well their personal lives. However, does the type of activities portrayed in vlogs differ from activities developers in the industry perform? Would developers at a software company prefer to show activities to different extents if they were asked to share about their day through vlogs? To answer these questions, we analyzed 130 vlogs by software developers on YouTube and conducted a survey with 335 software developers at a large software company. We found that although vlogs present traditional development activities such as coding and code peripheral activities (11%), they also prominently feature wellness and lifestyle related activities (47.3%) that have not been reflected in previous software engineering literature. We also found that developers at the software company were inclined to share more non-coding tasks (e.g., personal projects, time spent with family and friends, and health) when asked to create a mock-up vlog to promote diversity. These findings demonstrate a shift in our understanding of how software developers are spending their time and find valuable to share publicly. We discuss how vlogs provide a more complete perspective of software development work and serve as a valuable source of data for empirical research.

Andrew Anderson, Jonathan Dodge, Amrita Sadarangani et al.

We present a user study to investigate the impact of explanations on non-experts' understanding of reinforcement learning (RL) agents. We investigate both a common RL visualization, saliency maps (the focus of attention), and a more recent explanation type, reward-decomposition bars (predictions of future types of rewards). We designed a 124 participant, four-treatment experiment to compare participants' mental models of an RL agent in a simple Real-Time Strategy (RTS) game. Our results show that the combination of both saliency and reward bars were needed to achieve a statistically significant improvement in mental model score over the control. In addition, our qualitative analysis of the data reveals a number of effects for further study.