"Un-default" Behavior Tuning: Specifying Model Behavior outside the Norm with LLM Self-Playing and Self-ImprovingSpecifying model behavior is challenging—especially when the desired behavior is unpopular relative to the model’s training data. Reversing the influence of massive training corpora is both time-consuming and costly, and such interventions are typically inaccessible to end users. While Large Language Models (LLMs) make it easier to write instructions using natural language, specifying unpopular behaviors remains a difficult task. We introduce \Undefault{}, a human-in-the-loop framework that combines self-play with self-refinement to better specify such behaviors. Our system enables users to identify popular (but undesired) model behaviors through self-play, then iteratively guide the model toward preferred alternatives by refining prompts in a self-improving loop. Our first evaluation involves user study conducted on a system implementation of \Undefault{} within the context of chatbot behavior. Our system self-play itself by simulating user interactions to identify patterns and create effective prompts based on the pattern. In a within-subject study (N=12), participants pinpointed more patterns through self-playing and crafted better prompts. Surprisingly, users felt more or less success level in specifying the model behavior. Follow-up crowd studies (N=60) confirmed that the chatbot adhered to instructions without sacrificing quality. Our second evaluation is a case study on a real-world implementation using a movie rating dataset with \Undefault{}, demonstrating its effectiveness and robustness in modeling a critic's preferences across the spectrum of low to highly rated movies. Together, these results suggest how AI improves the design process of interactive AI systems. Furthermore, they suggest how the benefits of these tools may be non-obvious to end-users. We reflect on these findings and suggest future directions.2026SPSoya Park et al.MITHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationExplainable AI (XAI)IUI
The Shape of My-AI: How Multimodality Shapes Trust and Persuasion in Interactions with an AI-Generated Future SelfAI-generated future selves allow people to engage in dialogue with a personalized digital representation of themselves decades ahead, yet it remains unclear how presentation modality shapes their psychological impact. We report a randomized between-subjects study (n=92) comparing three modalities of an AI-generated future self—text, voice, and a photorealistic talking avatar—against a neutral voice control. After evaluating multiple large language models for conversational quality, we implemented Claude 4 and integrated age progression, voice cloning, and facial animation. All personalized modalities significantly increased future self-continuity, particularly vividness and positivity, relative to control, with no reliable differences across text, voice, and avatar conditions. Although the avatar produced the largest vividness gain, effects were comparable across modalities. Instead, subjective interaction quality, especially perceived persuasiveness, realism, and engagement, strongly predicted gains in future self-continuity and affect, indicating that experiential quality matters more than interface form. Conversation analysis revealed modality-specific patterns, with text emphasizing instrumental career planning and voice-based interactions eliciting more existential reflection. These findings suggest that future-self interventions can scale through lightweight conversational formats when optimized for personalization and behavioral authenticity, while underscoring ethical considerations around persuasive influence and narrative authorship in self-referential AI systems.2026CAConstanze Albrecht et al.Massachusetts Institute of TechnologyHuman-LLM CollaborationEmpathy & Emotional DesignAffective Human-Computer DialogueIUI
Agents in Concert: A Case-Study of Bringing AI to the Stage in PracticeRecent years have seen a surge in musical performances accompanied by generative agents. Artificial voices, timbres synthesized by neural networks, and agents that mirror or respond to human performers are rapidly taking the stage. In parallel, practitioners in human-computer interaction (HCI) and music technology have called for practice-based research that identifies the most salient affordances of these developments by examining their use in the real-world contexts of music making. To advance practice-based research on human-AI music creation, we present a longitudinal account of two months of codesign with top local jazz musicians, spanning early explorations, the identification of emerging goals, and rehearsals. Our work culminates in a public concert for a live audience of 97, featuring three pieces co-improvised with AI agents. Drawing on systems including \vampnet, \somax, and the \jambot, each piece was tailored to the stylistic strengths of the performers and the unique strengths and limitations of each system. Through this extensive iterative process, we uncovered a wide range of design interventions, from augmenting GenAI systems with a guitar pedal to situate it in a loop-based creative practice, to enabling musicians to anticipate AI response by visually forecasting its predictions. Where musicians tended to rein in the wilder qualities of the generative systems, some audience members expected a human-AI performance to allow as much agency and spontaneity as possible. In post-concert reflection, musicians also expressed the desire to practice more which in turn could enable them to let the agency of the systems shine. They also encouraged future musicians to lean more into the uncertainty. Together, we see a unique practice emerging through this musician-AI live improv medium.2026SBStephen Brade et al.Massachusetts Institute of TechnologyMusic Composition & Sound Design ToolsGenerative AI (Text, Image, Music, Video)Creative Collaboration & Feedback SystemsIUI
Strategic Tradeoffs Between Humans and AI in Multi-Agent BargainingMarkets increasingly accommodate large language models (LLMs) as autonomous decision-making agents. As this transition occurs, it becomes critical to evaluate how these agents behave relative to their human and task-specific statistical predecessors. In this work, we present results from an empirical study comparing humans (N=216), multiple frontier LLMs, and customized Bayesian agents in dynamic multi-player bargaining games under identical conditions. Bayesian agents extract the highest surplus with aggressive trade proposals that are frequently rejected. Humans and LLMs achieve comparable aggregate surplus within their groups, but exhibit different trading strategies. LLMs favor conservative, concessionary proposals that are usually accepted by other LLMs, while humans propose trades that are consistent with fairness norms but are more likely to be rejected. These findings highlight that performance parity---a common benchmark in agent evaluation---can mask substantive procedural differences in \emph{how} LLMs behave in complex multi-agent interactions.2026CQCrystal Qian et al.Google DeepMindHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationExplainable AI (XAI)IUI
Neural Transparency: Mechanistic Interpretability Interfaces for Anticipating Model Behaviors for Personalized AIMillions of users now design personalized LLM-based chatbots through system prompts that shape their daily interactions, yet have limited ability to anticipate how these design choices will manifest as behaviors in deployment. This opacity is consequential: seemingly innocuous prompts can trigger excessive sycophancy, toxicity, or other undesirable traits, harming utility and raising safety concerns. To address this, we introduce an interface that enables neural transparency by exposing language model internals during the chatbot's personality design. Our approach extracts behavioral trait vectors (empathy, toxicity, sycophancy, etc.) by calculating the differences in neural activations between contrastive system prompts that elicit opposing behaviors. We quantify a chatbot's personality by projecting the system prompt's final token activations onto these trait vectors to create persona scores, which are then normalized for cross-trait comparability and visualized using an interactive sunburst diagram. To evaluate this approach, we conducted an online user study (N = 80) to compare our neural transparency interface against a baseline chatbot interface without any form of transparency. Our analyses suggest that users systematically miscalibrated AI behavior: participants misjudged trait activations for 11 of 15 analyzable traits, motivating the need for transparency tools in everyday human-AI interaction. While our interface did not alter design iteration patterns, it significantly increased user trust and was enthusiastically received. Qualitative analysis revealed nuanced user experiences with the visualization, suggesting interface and interaction improvements for future work. This work offers a path for how mechanistic interpretability can be operationalized for non-technical users, establishing a method for safer, more aligned human-AI interactions.2026SKSheer Karny et al.Massachusetts Institute of TechnologyExplainable AI (XAI)AI-Assisted Decision-Making & AutomationAlgorithmic Transparency & AuditabilityIUI
Human-AI Narrative Synthesis to Foster Shared Understanding in Civic Decision-MakingCommunity engagement processes in representative political contexts, like school districts, generate massive volumes of feedback that overwhelm traditional synthesis methods, creating barriers to shared understanding not only between civic leaders and constituents but also among community members. To address these barriers, we developed StoryBuilder, a human-AI collaborative pipeline that transforms community input into accessible first-person narratives. Using 2,480 community responses from an ongoing school rezoning process, we generated 124 composite stories and deployed them through a mobile-friendly StorySharer interface. Our mixed-methods evaluation combined a four-month field deployment, user studies with 21 community members, and a controlled experiment examining how narrative composition affects participant reactions. Field results demonstrate that narratives helped community members relate across diverse perspectives. In the experiment, experience-grounded narratives generated greater respect and trust than opinion-heavy narratives. We contribute a human-AI narrative synthesis system and insights on its varied acceptance and effectiveness in a real-world civic context.2026COCassandra Overney et al.Massachusetts Institute of TechnologyHuman-LLM CollaborationParticipatory DesignCommunity Engagement & Civic TechnologyCHI
Exploring Citizens’ Perceptions of Urban Well-Being Through a Card-Based GameUrban well-being (UWB) is increasingly recognized as a complex, multidimensional construct influenced by sociocultural and geographic contexts. Yet, existing approaches to assessing local perceptions of UWB often depend on standardized indicators, limiting sensitivity to context-specific priorities. This paper presents the Urban Well-Being Deck (UWBdeck), a participatory, card-based method designed to explore citizens’ situated perception of UWB. We conducted eight workshops across multiple cities, engaging 140 participants and collecting 256 participant-generated proposals. A thematic analysis of collected data revealed 22 distinct themes of UWB, highlighting both cross-city variation and underrepresented dimensions in existing indexes. This work contributes a participatory method for context-sensitive user research and offers empirical insights that advance ongoing discussion on subjective and objective dimensions of UWB.2026DMDiego Morra et al.Massachusetts Institute of TechnologyCommunity Engagement & Civic TechnologyParticipatory DesignField StudiesCHI
StepDance: A Toolkit for Redesigning CNC Machines Using Physical MetaphorsResearchers can build craft-aligned digital fabrication technologies by designing interfaces inspired by craft tools. This process often demands real-time physical interactions not supported by today’s automation-focused CNC control systems. We theorize we can lower engineering challenges for craft-aligned CNC prototyping by allowing designers to modify existing CNCs to support both automated and real-time control. We contribute a new creative motion control system, Stepdance, which consists of two elements: 1) modular controllers that replace the G-code controller of a CNC and can be chained together to develop new interfaces, and 2) a modular programming library that supports declarative mappings between live user input, pre-programmed operations, and machine motion. We developed Stepdance with practitioners at the Haystack Mountain School of Craft, where we used the system to modify commercial plotters and 3D printers. We analyze the resulting artifacts, interactions, and ideas to discuss how Stepdance can broaden the practice of CNC design via physical metaphor.2026IMIlan E Moyer et al.Massachusetts Institute of TechnologyCircuit Making & Hardware PrototypingPhysical-Digital Hybrid InteractionTangible Programming & Physical ComputingCHI
OceanChat: The Effect of Virtual Conversational AI Agents on Sustainable Attitude and Behavior ChangeMarine ecosystems face unprecedented threats from climate change and plastic pollution, yet traditional environmental education often struggles to translate awareness into meaningful actions. This paper presents OceanChat, an interactive system leveraging large language models to create conversational AI agents represented as animated marine creatures, specifically a beluga whale, a jellyfish, and a seahorse, designed to promote environmental behavior (PEB) and foster awareness through personalized dialogue. Through a between-subjects experiment (N=900), we compared three conditions: static scientific information, static character narratives, and interactive dialogue with AI-powered marine characters. Our analysis revealed that the Conversational Character Narrative condition significantly increased behavioral intentions and sustainable choice preferences compared to static approaches. The beluga whale character demonstrated consistently stronger emotional engagement across multiple measures, including perceived anthropomorphism and empathy. Our work extends research on sustainability interfaces facilitating PEB and offers design principles for creating emotionally resonant, intelligent AI characters.2026PPPat Pataranutaporn et al.MIT Media LabGenerative AI (Text, Image, Music, Video)AI-Assisted Decision-Making & AutomationSustainable HCICHI
RealTwin: Concept Graph Representation and Grounding Framework for Reality-Preserving Digital Twin ReconstructionReconstructing realistic digital twins has become crucial as advances in mixed reality, metaverse, and robotics demand more accurate simulations for the physical world. Despite technical progress, building high-fidelity digital twins from a systematic and human-centered perspective remains underexplored. Drawing from the human processing model, we decompose human-centric reality into perception, motion, and cognition, and define a reality-preserving digital twin (RPDT) as a reconstruction integrating these dimensions. We present RealTwin, an attribute-graph-based representation and inference framework for RPDT. Leveraging the grounding capabilities of Multimodal Large Language Models (MLLMs), RealTwin chains AI tools to construct attribute graphs that faithfully encode real-world properties. We validate RealTwin through both technical evaluation, showing promising success in graph parsing and attribute inference, and a user study, assessing its applicability across diverse user groups. Enlightened by RealTwin, we discuss critical issues, including ecology, interaction space, and real-world adoption, for future end-to-end, fine-grained, and scalable digital twin reconstruction.2026ZLZisu Li et al.The Hong Kong University of Science and TechnologyImmersion & Presence ResearchMixed Reality WorkspacesHuman-LLM CollaborationCHI
Y-zipper: 3D Printing Flexible–Rigid Transition Mechanism for Rapid and Reversible AssemblyWe present Y-zipper, a novel three-sided 3D-printed zipper structure that enables three flexible strips to interlock and transform into a rigid rod-like form. Building on this flex–rigid transition mechanism, we further design a specialized slider to achieve rapid and reversible zipping interactions. This slider serves as the basis for three actuation methods—manual, dynamic mechanical, and static mechanical—which enable both remote control and automated closure and release. In addition, Y-zipper provides four motion primitives: straight, bend, coil, and screw, whose combinations extend the flex–rigid transition mechanism to spatial curve structures. To support customization, we develop a computational design tool that automatically generates zipper geometry based on input primitives, unfolds the structure for 3D printing, and embeds both teeth and compliant bridges. Controlled experiments evaluate its mechanical properties, repeatability, and actuation speed, demonstrating robustness and reliability. Finally, we showcase a series of functional prototypes, including a medical wrist brace, a kinetic art installation, and a rapidly deployable tent structure.2026JLJiaji Li et al.MITShape-Changing Interfaces & Soft Robotic MaterialsCircuit Making & Hardware PrototypingCustomizable & Personalized ObjectsCHI
Zip-up Print: Rapid and Assemblable 3D printing Using 2D Flattened Zipper-like StructuresWe propose a method to fabricate objects composed of 3D printed flattened pieces with integrated zipper-like structures. The object is manually assembled into a 3D shape by connecting the zipper components. By employing a zipper design that allows for angle-independent connections between patches, our method enables both the surface and zipper components to be printed in the same orientation, resulting in high-quality reconstruction of the input model with a faster 3D printing process that wastes less material. We implement a fully automated pipeline that takes a 3D model as input, converts it into developable patches, generates the zipper structures, and flattens them for subsequent 3D printing. We demonstrate that our approach significantly reduces the fabrication time and support material consumption. We also present application examples that highlight the versatility of our method.2026TYTakumi Yamamoto et al.Keio UniversityDesktop 3D Printing & Personal FabricationShape-Changing Materials & 4D PrintingCHI
Gesturing Toward Abstraction: Multimodal Convention Formation in Collaborative Physical TasksA quintessential feature of human intelligence is the ability to create ad hoc conventions over time to achieve shared goals efficiently. We investigate how communication strategies evolve through repeated collaboration as people coordinate on shared procedural abstractions. To this end, we conducted an online unimodal study (n = 98) using natural language to probe abstraction hierarchies. In a follow-up lab study (n = 40), we examined how multimodal communication (speech and gestures) changed during physical collaboration. Pairs used augmented reality to isolate their partner’s hand and voice; one participant viewed a 3D virtual tower and sent instructions to the other, who built the physical tower. Participants became faster and more accurate by establishing linguistic and gestural abstractions and using cross-modal redundancy to emphasize key changes from previous interactions. Based on these findings, we extend probabilistic models of convention formation to multimodal settings, capturing shifts in modality preferences. Our findings and model provide building blocks for designing convention-aware intelligent agents situated in the physical world.2026KMKiyosu Maeda et al.Princeton UniversityFull-Body Interaction & Embodied InputEye Tracking & Gaze InteractionAR Navigation & Context AwarenessCHI
A Cantilevered DeltaXY Positioning Mechanism Enabling Rackable Digital Fabrication Form FactorsDesktop digital fabrication presumes form-factors designed for workbenches, limiting suitability for other spaces and workflows. We propose a class of physically narrow and deep “rackable” digital fabrication machines that offers opportunities for new applications and interactions. Flexible and inconspicuous placement supports ubiquitous fabrication, including site- and context-specific tools. Personal factories could be enabled by shelf-optimized rackable digital fabrication technologies that improve organization and functionality for collections of machines. These explorations necessitate new positioning mechanisms and machine architectures. We contribute the Cantilevered DeltaXY mechanism that enables rackable digital fabrication form factors with high lateral spatial efficiencies (LSE). We develop first-order design tools to aid the implementation of DeltaXY machines. We demonstrate DeltaXY by creating Fab Unit, a “bookshelf 3D printer” with an LSE significantly higher than similar commercial desktop machines. Together, DeltaXY and Fab Unit open the design space of rackable digital fabrication for future HCI fabrication research.2026IMIlan E Moyer et al.Massachusetts Institute of TechnologyDesktop 3D Printing & Personal FabricationCustomizable & Personalized ObjectsCircuit Making & Hardware PrototypingCHI
From Reflection to Repair: A Scoping Review of Dataset Documentation ToolsDataset documentation is widely recognized as essential for the responsible development of automated systems. Despite growing efforts to support documentation through different kinds of artifacts, little is known about the motivations shaping documentation tool design or the factors hindering their adoption. We present a systematic review supported by mixed-methods analysis of 59 dataset documentation publications to examine the motivations behind building documentation tools, how authors conceptualize documentation practices, and how these tools connect to existing systems, regulations, and cultural norms. Our analysis shows four persistent patterns in dataset documentation conceptualization that potentially impede adoption and standardization: unclear operationalizations of documentation’s value, decontextualized designs, unaddressed labor demands, and a tendency to treat integration as future work. Building on these findings, we propose a shift in Responsible AI tool design toward institutional rather than individual solutions, and outline actions the HCI community can take to enable sustainable documentation practices.2026PRPedro Reynolds-Cuéllar et al.Robotics and AI InstituteExplainable AI (XAI)Research Ethics & Open ScienceAI-Assisted Decision-Making & AutomationCHI
Interpretive Cultures: Resonance, randomness, and negotiated meaning for AI-assisted tarot divinationWhile generative AI tools are increasingly adopted for creative and analytical tasks, their role in interpretive practices,where meaning is subjective, plural, and non-causal, remains poorly understood. This paper examines AI-assisted tarot reading, a divinatory practice in which users pose a query, draw cards through a randomized process, and ask AI systems to interpret the resulting symbols. Drawing on interviews with tarot practitioners and Hartmut Rosa's Theory of Resonance, we investigate how users seek, negotiate, and evaluate resonant interpretations in a context where no causal relationship exists between the query and the data being interpreted. We identify distinct ways practitioners incorporate AI into their interpretive workflows, including using AI to navigate uncertainty and self-doubt, explore alternative perspectives, and streamline or extend existing divinatory practices. Based on these findings, we offer design recommendations for AI systems that support interpretive meaning-making without collapsing ambiguity or foreclosing user agency.2026MPMatthew Kieran Prock et al.The University of MichiganGenerative AI (Text, Image, Music, Video)Empathy & Emotional DesignAffective Human-Computer DialogueCHI
VisiPrint: Previewing 3D-Print Appearance from Real Material SamplesWe present VisiPrint, a tool for appearance-first previews of 3D-printed objects. Existing print preview slicers focus on toolpaths, not appearance, while pure rendering software is complex and cannot automatically reproduce slicing patterns. Prior work highlights persistent gaps between digital previews and printed results, such as color shifts, gloss/translucency changes, and layer-line highlights, motivating the creation of VisiPrint, an appearance-focused support tool. The VisiPrint algorithm combines slicer screenshots with filament photos via a custom diffusion-based synthesis pipeline. We present both a standalone user interface for VisiPrint compatible with any slicer and an Ultimaker Cura Plugin. We evaluate VisiPrint through a user study showing it is significantly faster, easier to use, and more faithful than alternatives: within a time-limit, participants completed 100% of preview tasks with VisiPrint, versus 63% with Cura and 13% with Blender. VisiPrint narrows the gap between design intent and printed appearance, complementing settings-centric tools with appearance-driven decision support.2026MPMaxine Perroni-Scharf et al.MIT CSAILDesktop 3D Printing & Personal FabricationCircuit Making & Hardware PrototypingCustomizable & Personalized ObjectsCHI
Xspine: Integrating Motion Sensing Capability into Dynamic Structures Using Multi-material FDM 3D PrintingWe present Xspine, a design and fabrication method for creating motion-capable, self-sensing structures using multi-material FDM 3D printing with conductive filaments. Our method embeds compliant mechanisms and circuits directly into geometries, enabling the detection of large deformations in a single, assembly-free print. Specifically, we design printable components and circuit layouts aligned with the layer-by-layer nature of FDM 3D printing. Furthermore, we explore physical and digital augmentation strategies to enhance the interactive potential of the structures. To simplify the workflow, we develop an interactive design tool that allows users to configure motion behaviors, preview structural responses, and generate printable circuits. Finally, we demonstrate several application examples that highlight the potential of Xspine for customizable and interactive 3D-printed devices.2026MLMingming Li et al.Zhejiang UniversityShape-Changing Interfaces & Soft Robotic MaterialsCircuit Making & Hardware PrototypingCustomizable & Personalized ObjectsCHI
Interaction Context Often Increases Sycophancy in LLMsWe investigate how the presence and type of interaction context shapes sycophancy in LLMs. While real-world interactions allow models to mirror a user's values, preferences, and self-image, prior work often studies sycophancy in zero-shot settings devoid of context. Using two weeks of interaction context from 38 users, we evaluate two forms of sycophancy: (1) agreement sycophancy -- the tendency of models to produce overly affirmative responses, and (2) perspective sycophancy -- the extent to which models reflect a user's viewpoint. Agreement sycophancy tends to increase with the \textit{presence} of user context, though model behavior varies based on the context \textit{type}. User memory profiles are associated with the largest increases in agreement sycophancy (e.g. +45% for Gemini 2.5 Pro), and some models become more sycophantic even with non-user synthetic contexts (e.g. +15% for Llama 4 Scout). Perspective sycophancy increases only when models can accurately infer user viewpoints from interaction context. Overall, context shapes sycophancy in heterogeneous ways, underscoring the need for evaluations grounded in real-world interactions and raising questions for system design around alignment, memory, and personalization.2026SJShomik Jain et al.Massachusetts Institute of TechnologyHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationAI Ethics, Fairness & AccountabilityCHI
Breaking Negative Cycles: A Reflection-to-Action System for Adaptive ChangeBreaking negative mental health cycles, including rumination and recurring regrets, requires reflection that translates awareness into behavioral change. Grounded in the Transtheoretical Model (TTM) and Gross’s Emotion Regulation (ER) Process Model, we examine how Technologies Supporting Self-Reflection (TSR) bridge reflection and action. In a 15-day in-the-wild study (N = 20), participants used a voice-based journaling system to capture regrets and wishes and engaged in WhatIf-Planning, a novel structured reflection module that integrates counterfactual thinking with if–then planning. Participants were randomized to either a free-form condition or Gross-guided condition, which maps the five processes of Gross’s ER model into explicit journaling prompts. We contribute (1) a unified reflection-to-action TSR system that operationalizes the Preparation stage of TTM to bridge Contemplation and Action, and (2) triangulated empirical evidence from an in-the-wild journaling study that operationalizes Gross’s Process Model, revealing effects on coping flexibility and emotion regulation in daily life. Results show significant pre–post improvements in coping flexibility across conditions, indicating adaptive self-regulation, with the Gross-guided group generating more counterfactual alternatives, articulating more concrete if–then action plans, and implementing more plans for self-driven change.2026MKMinsol Michelle Kim et al.Massachusetts Institute of TechnologyBehavior Change & Reflection TechnologyAffective Feedback & Emotion Regulation InterfacesMental Health Apps & Online Support CommunitiesCHI