Large Language Models are reshaping task automation, yet remain limited in complex, multi-step real-world tasks that require aligning with vague user intent and enabling dynamic user override. From a formative study with 12 participants, we found that end-users actively seek to shape task-oriented interfaces rather than relying on one-shot outputs. To address this, we introduce the human-agent co-generation paradigm, materialized in DuetUI. This LLM-empowered system unfolds alongside task progress through a bidirectional context loop—the agent scaffolds the interface by decomposing the task, while the user's direct manipulations implicitly steer the agent's next generation step. In a technical ablation study and a user study with 24 participants, DuetUI improved task efficiency and interface usability, supporting more seamless human-agent collaboration. Our contributions include the proposal of this novel paradigm, the design of a proof-of-concept DuetUI prototype embodying it, and empirical and technical insights from an initial evaluation of how this bidirectional loop may help align agents with human intent and inform future development.
Researchers have characterized contexts for information work that are supported by often-messy ecosystems of information systems. Although important infrastructures for information work, little is known about how these less-than-perfect systems affect the data that is managed. We present results of an interview study of these information ecosystems in the nonprofit context. We find that the quality of data is often systematically degraded in five ways: incomplete data, out-of-date data, “bulk” data, “anecdotal" numbers, and “garbage" data. These forms of degraded data result from informants having other, legitimate priorities in their work—each more important than managing data. We discuss how degraded data often has little effect on organizations’ existing data practices, but forecloses alternate possible uses moving forward. Finally, we reflect on how we might be able to address these challenges while still respecting the legitimacy of choosing other priorities and explore what it would mean to design for a data imagination.
This paper investigates how indie game developers in resource-constrained, creativity centric small teams approach the potential of generative AI as a teammate in their collaborative workflows. Through 15 interviews with indie developers, our findings suggest that developers believe current AI systems still lack key elements of independence and interdependence that define a teammate in small indie teams. At the same time, they envision other constructive and desirable ways in which future AI could meaningfully support their teamwork, which highlight that AI should complement rather than directly participate in or imitate human creativity and collaborative dynamics. This work extends prior HCI research on human-AI teaming and collaborative creativity by shifting attention toward more socially nuanced and spontaneous creative teams beyond instrumental teams or individual creators. We also propose two new directions to rethink more nuanced ways to design future generative AI to better support indie game developers and other small creative teams alike.
Conceptual data modeling is a central activity in data work, yet how such models are created remains understudied. While data attributes play a key role, modeling is also shaped by tasks, tools, developers’ prior experiences, and often unfolds collaboratively between diverse stakeholders. In this study, we invited 22 participants with varying expertise in pairs to collaboratively sketch conceptual data models. We captured screen recordings, their evolving sketches, and conversations. Through a mixed-methods approach combining thematic analysis of dialogue with an examination of model artifacts, we identify how communication and collaboration patterns influenced the process. Our findings reveal a range of collaborative strategies and representations, as well as distinct ways dialogue shaped the emergence and expression of shared conceptual models. These insights deepen understanding of Human-Data Interaction in collaborative data work and point to design opportunities for tools that better support communication, negotiation, and sensemaking of data.
Cross-disciplinary teams increasingly work with high-dimensional scientific datasets, yet fragmented toolchains and limited support for shared exploration hinder collaboration. Prior immersive visualization & analytics research has emphasized individual interaction, leaving open how multi-user collaboration can be supported at scale. To fill this gap, we conduct semi-structured interviews with 20 domain experts from diverse academic, government, and industry backgrounds. Using deductive–inductive hybrid thematic analysis, we identify four collaboration-focused themes: workflow challenges, adoption perceptions, prospective features, and anticipated usability and ethical risks. These findings show how current ecosystems disrupt coordination and shared understanding, while highlighting opportunities for effective multi-user engagement. Our study contributes empirical insights into collaboration practices for high-dimensional data visualization & analysis, offering design implications to enhance coordination, mutual awareness, and equitable participation in next-generation collaborative immersive platforms. These contributions point toward future environments enabling distributed, cross-device teamwork on high-dimensional scientific data.
While Augmented Reality (AR) promises to transform remote collaboration, many aspects remain underexplored, particularly where to place remote avatars in messy, everyday environments. Two mixed-methods within-subjects studies examined avatar placement preferences during cooperative (brainstorming) and competitive (negotiation) tasks between participant pairs, focusing on the influence of physical objects (chairs, box, tree) on user preferences. Results showed a strong preference for frontal or slightly off-centre avatar placements, independent of task type. Participants preferred avatar placements that mirrored real-life behaviour, with chairs inviting placements and the tree deterring them. Notably, the large and visually simple box elicited mixed reactions, being viewed alternately as an obstacle to avoid when placing avatars or as an inviting physical anchor for them, despite causing a clear physicality conflict. We term this the "(Anti-)Affordance Problem", highlighting the complexity of avatar placement within physical contexts, and the necessity for AR collaboration platforms to respond to real-world constraints, offering flexibility in avatar placements to accommodate diverse user preferences.
Artificial Intelligence (AI) is rapidly reshaping team collaboration in workplaces. Meanwhile, women only represent 22% of the global AI workforce, raising questions about whose perspectives drive AI design. This dual reality makes the stakes high: without critical attention, AI may entrench existing gendered dynamics; but with deliberate design, it may open new avenues for equity. Through in-depth interviews with 30 AI professionals (22 women), our work both confirms gendered challenges in male-dominated teams and offers a novel contribution: how those who directly experience these dynamics envision AI’s role in mitigating them. These practitioner-informed design visions reveal AI's potential of offering multi-level support and empowering women to navigate these teams, and its risks of reinforcing stereotypes and surveillance. We call on the HCI community to explore this emerging design space for equitable human-AI teaming while critically attending to gendered power dynamics.