human-computer interaction teams

Abstract: As robots and intelligent systems increasingly interact with humans, they must anticipate long sequences of future actions from only brief observations. While purely neural models typically require extensive observations to predict future actions and remain difficult to interpret and adjust, symbolic priors offer a structured alternative. To this end, we introduce SymAnt, a novel neuro-symbolic approach that unifies the expressiveness of neural networks with the structured reasoning of scene and knowledge graphs. Given a short video, SymAnt performs a joint-search over both graphs to construct a symbolic context, enabling unified reasoning over spatial relationships, object attributes, affordances, and associated actions. This symbolic context serves as a prior for an iterative refinement process that progressively generates and improves the sequence of future actions. By grounding predictions in symbolic context, SymAnt anticipates long-horizon actions from short observation windows in a parameter-efficient manner, while enabling users to directly edit the underlying knowledge to adjust predictions without further training. Through our experiments, we demonstrate the effectiveness of SymAnt, particularly when only short observation contexts are available, across five diverse datasets: Breakfast, 50 Salads, EPIC-KITCHENS, EGTEA Gaze+, and Assembly101.

Abstract: AI agents no longer act alone: they cooperate, coordinate and compete within and between groups, increasingly alongside humans who are in, on or out of ``the loop''. Whether such mixed groups can govern themselves is an open question about machine behaviour and human-computer interaction that needs a testbed for hybrid institutions. We present Megabike, a multi-agent simulator and serious game with interdependent arenas for common-pool resource management, cooperative survival and constitutional choice. Riders sharing a megabike must agree their arrangements for steering, pedalling and distributing resources, and their political regime, while negotiating with rival bikes and avoiding an existential threat. Every rule is enforced by the simulator, every action is logged, and any seat can be taken by a human or a machine, so governance can be studied under well-defined metrics. We demonstrate Megabike with forty sessions in which language-model riders with cooperative, lazy, subversive or opposing dispositions chose regimes, elected leaders and expelled one another. The results show that different combinations of agent archetypes could produce unexpected and counterproductive outcomes. We conclude with extensions for systematic experiments mixing human and computational agents.

Abstract: Omnimodal large language models (Omni-LLMs) are increasingly expected to assist users while observing continuous audiovisual streams. However, existing evaluations provide limited insight into how reliably these models connect stream evidence and user needs to timely, sustained interventions. We introduce \textbf{Omni-InteractScope}, a benchmark of 1,479 human-annotated, cross-reviewed interactions with 1,790 temporal targets across eight task groups. It links input modalities, event triggers, intervention types, and user needs, covering event-timed tool invocation and query-free assistance alongside notifications, warnings, state tracking, and continuous responses. Our experimental results on the proposed benchmark reveal three critical limitations of current Omni-LLMs: (1) \textbf{Timely tool invocation remains unreliable}: all 12 evaluated models produce an on-time first response for at most 40\% of tool-invocation targets, and joint timing-and-tool accuracy reaches at most 32.0\%. (2) \textbf{Query-free reminders remain poorly covered}: all evaluated models achieve lower timely reference-target recall on query-free reminders than on explicit questions, measured on their respective target sets. Reminder recall never exceeds 23.0\%. (3) \textbf{End-to-end guidance coverage remains limited}: nine models cover none of the 26 guidance sequences in full, and the remaining three cover only one to three sequences each. Targeted controls further show that tool-invocation scores alone do not establish event grounding and that reminder coverage depends on response history and frequency. These findings motivate developing and evaluating streaming assistants for timely action, initiative under general user needs, and coverage of entire response sequences.

Abstract: In agent workflows, a brief human instruction can alter subsequent search, reasoning, and tool execution, yet final task outcomes rarely reveal the contribution of an individual instruction to human-agent joint production. We propose a Human Turn-level framework for measuring human contributions, using task quality and token usage to characterize the collaborative production outcomes associated with each human instruction. We compare continuations from adjacent checkpoints under matched models, budgets, and termination conditions, and assess stability through repeated measurements. Across four archived and simulated interaction settings, we conduct 29,124 comparisons between predicted and observed directions of quality and computational change at both Human Turn and session levels. On HumanBench, we obtain 506 valid comparisons across 85 real-user scenarios. Results show that quality and computational changes associated with human guidance depend on task conditions. Expert-style and ordinary-style guidance differ in information provision, correction, and repetition. Even when task quality after continuation is identical, remaining computational demands can differ substantially across checkpoints. Our framework advances the measurement of human labor contributions from tasks and sessions to individual turns, providing a repeatable, fine-grained foundation for studying human contributions and their allocation in human-agent joint production.

Abstract: EEG-eye movement emotion recognition can support mental-health monitoring and adaptive interfaces, but most fusion models treat the two signals as parallel features, classify each window alone and degrade when eye data are lost. We propose the Context-Aware Topological Distillation Graph Network (CTD-GNet), in which EEG features form the nodes of an electrode graph, and eye movements and preceding windows modulate its edges, nodes and readout, the edges through electrode-specific prompts. Bidirectional distillation of the graph, node relations and predictions couples this multimodal teacher with a single-window EEG student, improving the teacher and aligning the two graphs 35-fold. In leave-one-subject-out tests on SEED and SEED-IV against 17 baselines, CTD-GNet has the highest mean accuracy, 84.91% and 76.17%, significantly above 9 and 6 of 13 eye-based baselines after Holm correction, the smallest cross-subject spread among multimodal models and the highest worst-subject accuracy. It is the most accurate model at every rate of missing eye data on SEED-IV, keeps 66.1% on SEED without eye data, more than any EEG-only model, and stays 9.3 and 11.9 points above all single-window fusion models under eye-movement noise of one standard deviation. On film clips unseen in training, it leads on SEED with 77.1%.

Abstract: Vision-language-action (VLA) policies must ground language instructions in the right objects, contact points, and target locations for reliable manipulation. Human gaze provides a natural signal for this grounding, but existing robot datasets rarely pair real human gaze with the multi-camera observations used by modern policies. We introduce GazeAct, a gaze-equipped real-robot manipulation dataset and training framework for multi-view VLA policies. Using screen-based teleoperation, we record gaze directly on the robot camera feed viewed by the operator, yielding 12000 demonstrations across 40 tasks and five domains with 300 Hz gaze synchronized with images, proprioception, and actions. We transfer gaze from the operator-selected view to the remaining synchronized cameras, producing spatially consistent targets across the policy's visual inputs. We then use these gaze-derived targets to regularize language-conditioned visual attention during training while keeping inference gaze-free. Across 18 real-world tasks, our method improves average success from 48.1% to 58.6%, with gains up to 25 percentage points. Ablations show that gaze works best as soft supervision and that real human gaze provides stronger attention targets than generic attention-shaping priors. The gains transfer to LIBERO-Spatial (57.9% to 69.6%) and different VLA backbones.These results show that gaze collected in the robot observation space can provide effective training-time targets for improving visual grounding in VLA manipulation.

Abstract: Computer-use agents (CUAs) are commonly evaluated against fixed task specifications, but users may omit requirements, add constraints, or require approval before consequential actions. Completing the benchmark instruction does not guarantee serving the person who asked. We introduce PersonaCUABench, which holds 74 tasks across 11 domains fixed and varies only who asks. A persona agent, playing one of 44 personas in four groups, rewrites each instruction in that user's own words and answers the CUA's questions, while the CUA sees only the dialogue, never the profile or the original instruction. Across nine backbones (3,330 runs), mean task completion falls by 36% relative to the original requests. The losses have two sources: in 59.9% of cells that lose points, the agent engages with what the user asked for and still gets it wrong, and 11.2% are stops the user requested that the checker scores as failures. Even when a persona request is fully completed, the user gives the top satisfaction rating in only 62.7% of cells, against 84.1% for the original requests. CUAs should therefore be evaluated not only on task completion but also on whether they meet the requirements and constraints users express. Code: https://github.com/MatrAIx-ai/PersonaCUABench. Project page: https://matraix.ai/research/persona-cua.html

Abstract: Large language models (LLMs) reproduce a human working-memory signature on the N-back task: performance declines as memory load N increases, and errors concentrate on lures, items that repeat the current stimulus at a neighboring but incorrect serial position. We ask whether this shared behavioral signature reflects a shared mechanism. The literature offers two candidate accounts. On a precision account, softmax attention cannot concentrate its weight on the item N positions back, so adjacent items are retrieved instead. On an interference account, attention selects the correct item, but the content of competing items corrupts the decision. We dissociate these accounts in 15 pretrained LLMs from three families (70 million to 14 billion parameters) and in minimal transformers trained from scratch. In all three families, the attention heads that carry the task attend to the target on most missed matches. Linear probes decode near-perfectly whether a matching letter occurred at the target or at a lure position, even on trials where the model false-alarms to the lure, although its own decisions barely distinguish the two. Sharpening attention yields negligible gains, whereas suppressing the letter identity of non-target items improves sensitivity (d') in nearly all model-by-load conditions. In larger models the task is spread over more heads, and removing interference helps more. In minimal transformers, precision limits a single attention layer, but a second layer removes this constraint. The working-memory limit of LLMs therefore lies in reading items out under interference rather than in locating them, consistent with interference accounts of human working memory.

Abstract: Systematic literature reviews follow the PRISMA process of screening scholarly works and extracting evidence from included studies. Deep research tools based on large language models (LLMs) can search and summarize the literature but do not record which studies are excluded or where extracted values come from. We develop a conversational multi-agent system called ReviewPilot that automates the PRISMA workflow from paper screening to information extraction. Researchers can tailor criteria and schemas through dialogue, and a project memory keeps their configurations and revisions. Each screening decision comes with its reason and each extracted value with its source. On epidemiological expert reviews, ReviewPilot outperforms a single-prompt LLM and AgentSLR across three LLMs, with screening macro F1 up to 0.776 and extraction F1 up to 0.788. Case studies in biomedicine and human-computer interaction show close agreement with human annotators, suggesting that ReviewPilot can transfer beyond epidemiology. Code is available at https://github.com/ZhaoqianXue/ReviewPilot.

Abstract: Omni-modal large language models (Omni-LLMs) extend LLMs to process heterogeneous modalities (e.g., vision, audio, and IMU signals) alongside language, enabling a broader range of multimodal tasks. Yet we find that these models often fail to express information from non-textual modalities even when they perceive it correctly. We term this disconnect knows-but-cannot-say. To investigate this phenomenon, we introduce Token-level Modality Alignment (TMA), which measures the alignment between individual modality representations and response representations. Our analysis shows that non-textual representations are less aligned with response tokens and receive less attention than textual representations, suggesting a potential explanation for uneven modality preferences. Motivated by these findings, we propose OmniCatalyst, a simple perceive $\rightarrow$ infer strategy that complements the original modality inputs with self-generated text, making non-textual knowledge more accessible. Across seven Omni-LLMs, OmniCatalyst consistently improves performance (+7.29%) and yields more balanced modality preferences, with further gains from diverse training strategies. Further analysis demonstrates that the knowledge extraction phases increase TMA, suggesting that improved alignment makes non-textual information more accessible during response generation. In summary, our findings uncover how Omni-LLMs can better express modality knowledge through language. Our code is available at https://github.com/naver-ai/OmniLLM.

Abstract: Affective computing has progressed from categorical emotion recognition to open-ended affective analysis with large multimodal models. Yet affective science describes emotion as an unfolding process shaped by appraisal, regulation, and social interpretation, which remains underexplored computationally. We propose TRACE, a cognition-oriented framework that formalizes an affective episode through three interrelated stages: Condition, Affect, and Effect, integrating observable cues with cognitive factors such as internal stance and regulation of emotional display. Based on this formulation, TRACE-Bench evaluates multimodal models in real-world social scenes through five tasks spanning grounded affect recognition, regulation decoding, cause reasoning, effect reasoning, and full-chain reconstruction, with 3,746 structured question-answer pairs over 646 videos. A matched human-model comparison reveals a substantial performance gap, while affect-specialized models also generally lag behind general-purpose MLLMs. Model outputs show recurring failures, including treating displayed behavior as genuine feeling and fabricating unsupported events during long-chain generation. We further propose TRACER, a cognition-grounded structured reasoning method that couples each inference with explicit premises from factual observations, cognitive appraisals, and established upstream conclusions, forming a traceable graph of intermediate and target conclusions. TRACER outperforms all evaluated model baselines on each of the five tasks. Project page: https://cogaffc.github.io/TRACE

Abstract: Streaming video-language models must decide not only what to answer, but whether the evidence needed for the current question has arrived. Existing systems learn that decision as a separate trigger; we ask whether an unmodified model already computes it. We show that frozen VideoLLMs carry a linearly readable evidence-readiness signal, labelled from timestamped evidence rather than from model output. It decodes in all seven models of a shared byte-identical evaluation (AUROC 0.733-0.905 under the strictest not-ready sampling, where a fitted clock is near chance), and a probe fitted without any of a benchmark family's footage still reads that family. It is question-conditioned: on byte-identical windows, changing only the question reverses the readout on 66.1% of pairs, while every question-blind control is at chance by construction. The model can answer incorrectly and still encode readiness: AUROC remains 0.722 among wrong answers. Readiness also beats uncertainty estimators and their supervised combination on latency-matched answer selection, and tracks independent human judgments more closely than confidence. Released streaming triggers are also linear readouts, yet a trained trigger read on its own base model's activations is approximately orthogonal to readiness and decodes it far less accurately than a probe. We turn the readout into Readiness Gating, an answer-timing policy that improves accuracy by up to +9.75 pp at matched video duration with negligible computational overhead. How much it gains varies with the accuracy headroom the task makes available: across 26 configurations the gain tracks that headroom, and an intervention that moves it over identical pixels moves the gain with it.

Abstract: Electroencephalography (EEG) emotion recognition serves as a pivotal interface for human-computer interaction, yet the physiological variability across individuals complicates the already challenging task of fusing heterogeneous physiological signals (e.g., EEG and eye movements). However, most prevalent domain adaptation paradigms are tailored for unimodal scenarios, failing to address the heterogeneity of multimodal signals. Furthermore, they predominantly rely on feature-level alignment, overlooking the fundamental data-level discrepancy, which risks compromising fine-grained discriminative information during aggressive adaptation. To bridge these coupled gaps, we propose Instance-based Domain Expansion and Adversarial Learning (IDEAL), a unified framework that synergizes instance-level curriculum expansion with feature-level hierarchical adversarial alignment. IDEAL first introduces a multi-model collaborative screening mechanism, which propagates high-confidence target samples to explicitly bridge the distributional gap at the data level via a quantity-quality equilibrium strategy. We provide a theoretical analysis that this instance expansion strategy strictly tightens the upper bound of the target risk. Subsequently, a hierarchical adversarial network, augmented with the angular-contrastive constraints, progressively aligns representations from low-level statistics to high-level semantics while preserving class separability. Extensive experiments on four benchmark datasets demonstrate that IDEAL significantly outperforms state-of-the-art methods. To facilitate reproducibility and future research, our source code is publicly available at https://github.com/WY-BCI-Club/IDEAL.

Abstract: Neural codecs encode continuous signals into compact sequences of discrete tokens, providing an interface for efficient transmission, storage, and token-based sequence modeling. This paradigm has been widely adopted in modern speech and audio frameworks; however, the biosignal domain still lacks a neural codec designed specifically for low-bitrate streaming and generalization across diverse downstream tasks. We present MyoCodec, a streaming neural codec designed for electromyography (EMG). Inspired by recent neural audio codecs, MyoCodec combines causal Transformers with residual vector quantization to encode continuous EMG signals into different levels of EMG representations spanning from continuous latent features to discrete tokens operating at 50 Hz. Trained on twelve public EMG datasets, MyoCodec achieves favorable performance in both intrinsic codec quality and representative downstream tasks, including typing (emg2qwerty), hand-pose (emg2pose), speech decoding (emg2speech), and speech-to-EMG synthesis (speech2emg). Across these tasks, MyoCodec exhibits strong performance against prior models while providing a compact and causal EMG representation. During streaming inference, it requires compute time of only 0.482 ms for each 20 ms frame, enabling real-time streaming. Also, the discrete token representation provided by MyoCodec has the potential to support integration into language-model based approaches, creating a path toward LLM-based interactive systems, where tokenized EMG representations are directly processed into such language or speech models. Code and model weights are released.

Abstract: Visual emotion recognition commonly assumes that all evidence required for prediction is contained in the observed image or video. Yet the same visible reaction can convey different emotions depending on events beyond the input: tears, for example, may indicate grief or joy. We formulate Event-Grounded Emotion Recognition (EGER), where emotion recognition requires recovering the affect-determining event. We construct EGER-Bench, comprising 10,052 videos and 10,734 images across 11 emotions, two source domains, and four visual settings. A study with six annotators shows that event context raises human recognition accuracy from 33.96% to 72.08%, confirming that visual evidence alone is often insufficient. Semantic relevance alone does not solve EGER: a plausible event may imply the wrong emotion if its identity, focal-person role, relationship, or outcome is misinterpreted. We therefore propose AffectReveal, a tuning-free framework that first constructs and independently verifies evidence-grounded alternatives over these affect-critical factors. It then cross-checks the recovered event against face-masked in-media facts through bidirectional atomic evidence support, while retaining the original unmasked input for final prediction. Across three downstream models and four input settings, AffectReveal yields average UAR gains of 5.26--10.53 points. For three fine-tunable models, it also enables untuned models to outperform their fine-tuned visual-only counterparts in all 12 accuracy comparisons, without updating downstream parameters.

Abstract: Sign data collection is costly, and teleoperation scales poorly, motivating reuse of large video corpora. Humanoid signing requires converting video-derived human motion into robot trajectories while preserving linguistic motion cues. Errors from fitting, human-motion repair, retargeting, robot geometry repair, and control are hard to separate from the final trajectory alone. We introduce HumanoidCSL-20K, a dataset and benchmark of 20,648 sentence-level Chinese Sign Language sequences, each with four aligned versions: the source, the repaired human motion, the direct robot reference, and the geometry-repaired robot reference. Observation-supported local human-motion repair, full-robot geometry repair, and cross-representation provenance make each transformation traceable. Paired evaluations measure human-motion continuity and content preservation, robot-reference feasibility, and physical execution. A sign-specific kinematic-reference protocol scores handshape, location, palm orientation, and inter-hand relation over the full planned motion. Full-corpus results show fewer abnormal arm / hand steps and less inter-hand and hand-body penetration after repair. Control experiments separate reference learnability from curriculum effects, while component scores expose remaining execution errors.

Abstract: Electroencephalography (EEG)-based emotion recognition is important for affective computing and human-computer interaction, yet most existing methods divide a long trial into short segments and assign each segment the label of its source trial. Although this strategy increases the number of training samples, it reduces an evolving emotional response to a segment-level, coarse-grained, and static prediction problem. In reality, emotion may continuously emerge, intensify, weaken, and fluctuate as a stimulus unfolds, motivating the prediction of a time-aligned affective trajectory from the complete EEG trial. This task requires coordinated modeling of how spatial neural organization evolves throughout the trial and how local emotional fluctuations interact with longer-term trends. In this work, we formally define and systematically investigate continuous EEG emotion recognition as whole-trial affective trajectory prediction. We propose EAGER, an Evolving Affective Graph framework with Evidence Retrieval for continuous EEG emotion recognition. EAGER comprises two complementary modules: Affective State-guided Topology Evolution models the evolving spatial organization of EEG activity, while Multi-scale Temporal Evidence Retrieval integrates short-term fluctuations with longer-range temporal trends for time-aligned prediction. Experiments on MAHNOB-HCI, SEED-VII, and REFED show consistent gains in trajectory-tracking metrics over representative methods, with competitive pointwise errors.

Abstract: Social Proactive Intelligence (SPI) extends proactive assistance beyond task completeness to consider social appropriateness in diverse embodied scenarios. However, prior SPI research faces two key limitations. First, existing work focuses on static images, whereas dynamic videos provide crucial cues for inferring human states and needs, offering richer information than isolated images. Second, prior work often relies on free-form data collection pipelines, which fail to guarantee comprehensive coverage of diverse scenarios. To address these gaps, we introduce RobotEQ-Video, shifting the focus from image-centric to video-centric analysis. To ensure comprehensive video coverage, we construct a hierarchical world-state taxonomy organized into a four-level coarse-to-fine structure, comprising 6 domains, 20 dimensions, 142 level-1 attributes, and 816 level-2 attributes. The resulting benchmark comprises 2K+ videos with 100K+ human annotations and 16K+ labels for assessing behavior properness. Benchmark evaluation reveals that current systems remain unreliable and fall short of human performance. We further explore how world models can help tackle this task. This work advances SPI research from static images to dynamic videos and ensures more comprehensive scenario coverage during benchmarking.

Abstract: Understanding and modeling human intelligence are parallel goals shared by artificial intelligence (AI) and cognitive science. As AI systems grow increasingly capable, in what ways do model responses resemble human responses, and where do they systematically diverge? The sheer breadth and diversity of the tasks humans can perform and think about pose a challenge for scalable and rigorous comparison between humans and models. We introduce CogGym, a scalable, unified framework grounded in cognitive science for systematically comparing model and human behavior on matched experimental trials. CogGym uses a semi-automated, human-in-the-loop pipeline to standardize diverse experimental paradigms into a task-agnostic Experiment Markup Language (EML), enabling reproducible and faithful comparison at scale. For initial release, we curate and standardize 258 cognitive experiments from 100 papers that focuses on human commonsense reasoning, and evaluate 50 large language models against human responses. We find a clear scaling trend where larger and more recent AI models better reproduce human judgments. Yet AI models' improvement on such common reasoning tasks is considerably slower than the gains observed on formal-reasoning benchmarks like math and coding, and model--human fit remains well below human splithalf reliability ($R^2 = 0.93$ on text, $0.95$ on image, and $0.92$ on video) with the best models achieving $R^2 = 0.59$ on text, $0.58$ on image, and $0.43$ on video experiments. We intend for CogGym to provide a living evaluation framework that continually incorporates new cognitive science experiments to characterize where model behavior resembles human behavior, where it systematically diverges, and how those patterns change as models and experiments evolve.

Abstract: Chess has long served as a model domain for studying search, expertise, decision-making, and artificial intelligence. The emergence of large language models (LLMs) has renewed the relevance of chess as a controlled environment for investigating strategic reasoning and comparing human and artificial decision-making. We present a systematic mapping study of recent research spanning human players, classical chess engines, neural and reinforcement-learning systems, LLMs, and hybrid approaches. The final map comprises 84 core study families, classified according to agent type, strategic-reasoning stages, and evaluation dimensions. The map reveals a literature strongly concentrated on situation assessment, evaluation, and action selection, while explicit planning, explanation, metacognition, and human--AI collaboration remain less explored. LLM research places particular emphasis on state representation and generalization, whereas grounded explanation appears more frequently in hybrid approaches combining language models with engines, expert knowledge, or other external structures. Two distinctions emerge that the map aggregates rather than resolves: hybrid systems differ in where and when heterogeneous capabilities combine, and evaluations that show improved human performance do not thereby establish human--AI synergy. We propose both as extensions of the mapping framework. We argue that chess provides a useful bridge between cognitive and computational perspectives on strategic reasoning, and identify explicit planning, grounded and faithful explanation, metacognitive calibration, and human--AI complementarity as directions for future research.

Abstract: The mental model construct is widely used in HCI to refer to the knowledge structure people hold in order to reason about and interact with computing systems. Yet it is often operationalized intuitively: the construct is often used interchangeably with related concepts (e.g., folk theories, sensemaking) and methods of studying it (e.g., through elicitation) are many and diverse, with each method resting on distinct assumptions about what counts as a mental model. Generative and agentic AI systems may further complicate mental model formation and elicitation as such systems are opaque by design and increasingly act on users' behalf across files, applications, and on the web. Together, these challenges may hinder the commensurability of research on people's mental models of AI systems. The MM/AI workshop calls for a critical reassessment of how we understand and study mental models in human-AI interaction research. It aims to foster theoretical and methodological exchange on mental models in human-AI interaction, identify open challenges, and develop directions for future research. We invite short papers on users' or stakeholders' mental models of AI systems, particularly contributions that reflect on the conceptual and methodological foundations of the construct. The half-day workshop combines lightning talks, hands-on elicitation exercises, and structured discussions on key questions concerning the future of the mental model for human-AI interaction research.

Abstract: Embedding an intelligent agent in an existing application creates a persistent coordination problem: users can revise goals and manipulate shared objects while delegated execution continues. We argue that dependable integration requires an explicit correspondence between task-level interaction and application behavior. We introduce Agent-Integrated Software (AIS) as a software pattern combining a conventional core, direct interaction, and a built-in agent, and Intent-Level Interaction Abstraction (IIA) as the task semantics through which users inspect and control delegated work. An open transition-system model relates AIS execution to IIA states and events. Interaction contracts constrain this relation through task bindings, role-specific authority, control transitions, and outcome evidence; continuous assurance maintains scoped claims as their dependencies change. A compact disclosure contract and conditional propositions illustrate why local component validity is insufficient and how selected admission invariants can be separated from planning. Contrasting software domains expose the framework's assumptions and limits. This perspective develops a research agenda spanning application abstraction, development support, controlled execution, quality assessment, and human supervision, with the aim of making agent integration a maintainable software engineering discipline.

Abstract: Organizational design in the era of artificial intelligence requires experimental methods that can test how human-AI groups coordinate, delegate, and make decisions. Programmable platforms coordinate live human-to-human sessions or real-time human-AI chat, but researchers cannot easily declare experiment protocols in which AI participants both communicate and act on shared work within one auditable configuration. Here we introduce Pairit, an online platform that facilitates the design, testing, and deployment of experiments that test human-AI organizational designs and interventions. Through a single YAML configuration file, researchers declare an executable experiment graph (pages, routing, randomization, matchmaking, chat, shared workspaces, server-hosted agents, surveys, timers, and custom HTML components) and combine any number of humans and AI agents in live sessions. We have validated the feasibility of the platform through multiple live deployments, including peer-reviewed published studies, capturing high-resolution process traces of communication, negotiation, and collaborative work in live human-AI dyads. By representing complex interactive protocols as standardized, auditable configuration files, Pairit provides reusable infrastructure for specifying, deploying, and sharing live human-AI organizational experiments.