Human-in-the-Loop Evaluation

Road safety programs count the coded fields of police crash records, while the officer's narrative, which often records factors the fields omit, is rarely read. A safety office thus cannot tell how much its counts miss or where to review. This study develops and evaluates a system that joins both views of the 5,601,890 Texas crashes from 2017 to 2025 into population estimates with stated validity. An in-context tabular foundation model, Kumo Tabular, reads the coded record of every crash, a calibrated System One model, Jev, reads the narratives of two probability samples, and human judgments recalibrate its probabilities. A multiwave predict-then-debias estimator joins the three tiers, and a second human tier drawn with recorded probabilities checks the estimates by design. For hydroplaning, medical episodes, fatigue, animals, and phone use, the narrative documents more injury crashes than the coded field, 15,074 against 7,340 for phone use, and the human check agrees with all fifteen estimates within its margin. A re-read list ranked by Kumo Tabular finds confirmed discordance 7 to 58 times as often as random reading. At the planning cost of human coding, one further round of human judgments would cut the root mean square relative half-width from 22.0 to 16.2 percent, against 21.2 for reading every narrative. Two calibrated readers of different views, joined by a sampling design, give a safety office counts, a discordance map, a validated re-read list, and a reading budget, with Kumo Tabular reading the table at 15 times the speed of TabPFN 3.5.

Human-Robot Collaboration (HRC) can facilitate mass customisation in Industry 4.0, with Reinforcement Learning from Human Feedback (RLHF) representing a promising approach for developing safe AI-based robots. Practical challenges remain regarding safety during AI development, human feedback quality, and bidirectional human-robot adaptation. We conducted a scoping review of RLHF in HRC systems, mapping methods that address these challenges. Following PRISMA guidelines, we screened 199 records and included 20 peer-reviewed publications (2020-2025) spanning multiple HRC domains. To our knowledge, this is the first review focused on the bidirectional, closed-loop design of RLHF. Our review found multiple feedback modalities enabling data collection in various feedback formats. Collected data can be integrated at different stages of AI training, resulting in a multi-step development process. Pilot experiments are commonly used to evaluate HRC systems based on both human and robot metrics. To empirically test a key gap identified in the review, we conducted a between-subjects VR experiment comparing system- and user-initiated feedback on robot proxemic behaviour for safe navigation. Using Bayesian models, we analysed the relation between the collected feedback and safety metrics: psychological safety (post-experiment questionnaire) and physical safety (inverse time-to-collision). Results show that user-initiated feedback captures perceived safety better than system-initiated feedback, indicating that feedback timing directly affects feedback quality. Our review and experiment findings show that RLHF relies on appropriate feedback methods to ensure AI safety in HRC, and future RLHF research should prioritise realistic HRC experiments evaluating the effects of feedback collection methods on relevant human and robot metrics.

As agents take on long-horizon tasks, users shift from making individual decisions to overseeing autonomous execution. Yet the volume of agent activity and the fragmentation of supporting evidence make it difficult to determine which decisions warrant user verification. We study monitors that identify consequential decisions and locate evidence to help users assess their implications. We introduce AgentMonBench, a software-engineering benchmark comprising three subsets that cover two complementary dimensions: alignment between requirements and behavior, and awareness of consequential autonomous decisions for verification. To support these judgments, we propose the Evidence-Grounded Behavior Graph (EBG), a training-free method that groups source-linked evidence into behaviors and organizes their relationships into a graph. EBG presents task-oriented views of this graph to help monitors interpret behavior in context. Experiments across eight models show that EBG improves decision identification and evidence localization in most settings compared with direct access to the original context. Further experiments show that EBG's evidence-localization gains persist across input scales and hyperparameter settings, while real-world applications illustrate its practical value for human oversight.

Web agents are an important application of large language models, yet their evaluation often depends on rule based or language model evaluators that inspect only the final outcome. Human verification of task completion and detailed analysis of failed trajectories remain limited. We audit all 165 WebArena Lite tasks under six evaluation conditions built from GPT 5.5 and an untrained Qwen3.5 9B model. The audit retains the original score, corrects false negatives from the automatic evaluator, identifies the first consequential error, and examines progress across the trajectory. We also study a Memory and Analysis Support Mechanism (MASM), which maintains explicit execution state, and Guide Text, which provides task relevant procedural guidance. Across four GPT 5.5 settings, human review recovers 5.45 to 8.49 percentage points of success missed by the evaluator. With a 25 step budget, Guide Text raises corrected success with MASM from 34.55% to 38.18%. On the untrained Qwen3.5 9B model, MASM raises the evaluator score from 13.90% to 18.80%. Review of 102 failed GPT 5.5 trajectories reveals frequent scrolling loops, unfinished exploration, premature answers, invalid actions, and incomplete form workflows. Step level evidence further shows that substantial early progress can coexist with a final failure. These results show why final scores alone provide an incomplete account of web agent behavior and motivate human grounded, trajectory aware verification.

The ontology-based contextual AI evaluation (OB-CAIE) methodology was developed to address a lack of scientific rigor that arises from unclear testing coverage, to balance human expertise and automations, and to address a lack of reproducibility of AI evaluation testing environments. OB-CAIE strengthens the current state of AI evaluations by addressing the first step in the scientific method by clearly defining what will be tested. Two ontologies represent the tractable problem space in the OB-CAIE methodology: the Domain-Specific Ontology (DSO) and the Evaluation Process Ontology (EPO). The DSO is the what; the EPO is the how. An OB-CAIE problem space can be used for one or multiple AI evaluations. The OB-CAIE methodology allows for human judgment at specific points, in scientifically grounded ways, and in complex subject areas where human feedback is genuinely irreducible or machine irreplaceable. A key advantage of the OB-CAIE methodology is that failure points can be traced, visualized and analyzed within the canonical OB-CAIE methodology problem space.

Many LLM-as-a-judge evaluations score fixed outputs under a fixed task definition. Production multi-turn business agents instead require a maintained measurement system: correctness depends on business-specific facts and procedures, outcomes emerge across turns, and failures must be attributed to either agent capability or missing business knowledge before they are actionable. We present an integrated methodology spanning evaluation specification, modular LLM judges, intent-preserving user simulation, and human-in-the-loop governance. The specification defines conversation-level end states and actionable failure ownership. Atomic judges share versioned evidence and feed an explicit aggregation graph. The simulator is released only after task-preservation and stability checks. Independent human audits estimate measurement fidelity, renew tiered reference sets, and route disagreements to label correction, guideline revision, or judge improvement. Production studies show that system-level fidelity improved across repeated audits, that human reviewers and automated judges improved together under the shared feedback loop, and that their combined workflow had the strongest descriptive performance in both reported task-completion settings. Because the studies are observational and the human reference itself required revision, these findings demonstrate operational usefulness rather than causal or universal superiority. The contribution is a practical framework for making multi-turn agent measurement reliable, actionable, and maintainable as the evaluated system and its evidence evolve.

Effective robot assistance beyond narrow roles and repetitive tasks requires robots to be proactive - to decide what needs to be done rather than waiting to be told. While proactivity is increasingly explored, it lacks a unified formulation, and work in the domain is typically evaluated offline against static human models that cannot capture the effect of a robot's actions on the environment and the user's own behavior. We introduce a unified formalism for proactive robot assistance, organize it into three levels, and provide a framework to address the highest level of unprompted proactive assistance. We then show that offline evaluation overstates performance in this setting, and contribute a closed-loop evaluation with a human model that adapts to the robot. Finally, we present a method, GAP, that instantiates our framework, learning from passive observation to anticipate user goals and act. Under closed-loop evaluation, prior state-of-the-art methods collapse, in some cases adding more work than they save, while GAP remains robust and substantially outperforms them.

Large language models are increasingly used as inexpensive judges to evaluate outputs, label data, and assess whether a system meets a desired quality standard. Yet using AI judgments for formal statistical inference is fundamentally different from simply treating them as ground-truth labels: AI evaluations can be biased or noisy, and rigorous hypothesis testing requires explicit control of type-I and type-II errors. We study how to use AI judgments, together with selective human verification, to conduct a valid hypothesis test at minimum cost. We consider a population of items with hidden binary labels. After choosing a fixed pool of items, the decision maker can selectively query AI, send an item directly to a human, escalate an AI-scored item to a human after observing the AI report, or stop once sufficient evidence has accumulated. We derive an information-theoretic lower bound that captures the minimum cost of achieving prescribed testing errors and characterizes the value of AI information and human verification through a report-dependent information frontier. Motivated by this characterization, we develop SCALE, a sequential cost-aware policy that combines selective AI scoring with adaptive human escalation. SCALE is valid at finite sample sizes and matches the lower bound to first order as the target error probabilities vanish. We further extend the framework to an unknown AI-output model using paired AI-human pilot data. Numerically, SCALE approaches Human-only or AI-only testing when one source clearly dominates, while achieving its largest savings when inexpensive AI judgments and selective human verification are both valuable.

Multimodal large language models (MLLMs) are increasingly explored as interfaces for scientific image analysis, where a visual question-answering (VQA) response may be paired with a spatial output that guides a downstream stage. A supervisor reads the language answer, while a downstream workflow such as segmentation or region review consumes the point-set output. We call this transition from inspecting the answer to relying on its point action the language-to-action hand-off. A silent failure occurs when the answer is correct while the paired action misses annotated objects needed downstream, so answer-based oversight clears a region whose action is unreliable. We introduce Spatial Action Review, a visual analytics dashboard for auditing this failure mode in electron microscopy (EM) mitochondria analysis. It links paired answer-action records through an answer-action ledger, a task-by-dataset risk map, and an image-region audit view, connecting aggregate patterns to image evidence while an adjustable action-reliability gate supports re-audit. The review ends in a human-AI hand-off, where a supervisor records whether the action is accepted, escalated, held under a stricter gate, or flagged for model revision. Across 541 image regions from an EM-adapted Qwen3-VL case-study run, point actions fail the gate in 54.4% of records with a correct VQA response, and 27.4% of all records are silent failures. A correct answer is associated with only a 5.8-percentage-point higher probability of a reliable action, with a bootstrap interval spanning zero; the point-biserial correlation between answer correctness and object coverage is 0.061. This weak coupling persists across five model conditions on 753 matched image regions. Spatial Action Review makes answer-action mismatches visible and ties them to image evidence and a recorded decision before MLLM outputs enter autonomous scientific workflows.

Robots can recover from failures by asking bystanders for help, but effective human-in-the-loop recovery requires communication that accounts for differences in people's knowledge. Prior inverse-semantics work generates requests using a single listener model, leaving differences in listener knowledge untested. We introduce Listener Differences in Human-Robot Interaction (LD-HRI), a game, dataset, and benchmark that evaluates speakers through human listener performance. Our evaluation examines request properties, large language model (LLM) speakers, and inverse-semantics request-selection algorithms under controlled differences in listener information. The corpus contains 446 human-written requests and 1{,}302 listener trials. We additionally evaluated 24 frozen LLM-written requests with 70 human listeners across 560 trials. Novice success is descriptively higher with model-written requests across all four tasks, yet both request sources leave substantial expert--novice gaps, including 16 percentage points for LLM requests. LD-HRI makes these gaps measurable, providing a foundation for designing more robust communication in human-robot and human-agent interaction.

Frontier language models now produce professional deliverables that expert graders judge to match human work on a substantial share of economically valuable tasks, yet most enterprise GenAI initiatives fail to show a measurable business effect and a large fraction of agentic projects are expected to be cancelled. We argue that this is substantially a measurement problem: public benchmarks answer "what can the model do?", whereas a deployment decision requires "is this workflow fit, reliable, safe and worth scaling - here, on our data, under our controls?". We present EnterpriseVal, a use-case-level evaluation system that closes this gap. It comprises (i) a formal specification of the use case and of the frozen socio-technical configuration under test, model, prompts, retrieval, tools, guardrails and human oversight, with an autonomy level and consequence tier that jointly set the required evaluation intensity; (ii) a metric catalogue spanning fidelity, utility, efficiency, reliability, assurance and oversight; (iii) a grading protocol that scales blinded expert judgement with calibrated LLM-as-judge scoring through prediction-powered inference; (iv) a two-tier threshold gate, stated as an executable algorithm, that maps metric vectors with confidence bounds to REJECT/CONDITIONAL/SCALE decisions; and (v) a value-and-risk model in which the reviewer catch rate is a measured parameter. We report a pilot across three workflows in a global bank. In credit-memo drafting, human-graded citation precision reached 88% and hallucination rate 1.6% for the best model against gates of 70% and 5%; in procedure transformation, analyst refinement effort fell from an estimated 27.4 to 2.9 hours per document. We separate established results, documented pilot evidence, the proposed system and open hypotheses, and specify the experiments required for full validation

Machine learning (ML) has become the dominant approach for network traffic classification, achieving very high predictive performance. However, a model is only valuable if it learns semantically meaningful and trustworthy patterns rather than exploiting spurious correlations. Conventional evaluation practices predominantly assess predictive performance. Consequently, whether the model relies on semantically meaningful patterns remains unknown. To address these challenges, we adapt the knowledge generation framework for network traffic classification. The adapted framework combines data, ML models, explainability, visualization, and expert reasoning to support the iterative exploration, verification, and refinement of model behavior and data preprocessing. The framework is grounded in findings from the literature, benchmark dataset analyses, practical experience with XAI-based traffic classification, and expert feedback, providing practical guidance for semantic model validation. By complementing predictive performance with semantic validation and human expertise, the proposed framework supports the development of network traffic classification models that are not only accurate but also robust and trustworthy.

Hip exoskeletons may improve recovery from unexpected gait perturbations, yet personalizing assistance remains difficult because balance is multidimensional and human-in-the-loop experiments are small-sample and noisy. We present a participant-specific composite balance cost that integrates seven biomechanical sub-metrics spanning margin of stability, center-of-mass dynamics, and whole-body angular momentum. The sub-metrics are converted to direction-aligned, dimensionless cost features, and nonnegative fusion weights are learned on the simplex. Coupled with an empirical-Bayes hierarchical model, the learned-composite selector estimates each tested condition's posterior probability of being best, P(best), and a high-probability candidate set with size $K_{0.8}$. The framework was evaluated with three participants walking at 1.1 m/s during unilateral belt-slip perturbations across 46 hip-assistance conditions. In the full-budget analysis (B = 4 repeats per condition), the selector concentrated 80% of the posterior probability within 1 to 5 of 46 conditions, compared with 2 to 12 for equal-weight fusion and 4 to 37 for principal component analysis fusion. This smaller candidate set could shorten personalization experiments and limit participants' exposure to repeated perturbations in future studies. Selected-condition trials showed lower observed composite costs than no-torque trials, with nominal p < 0.05 for P2 and P3. Leave-one-repeat-out refits yielded positive mean held-out rank correlations for all participants and moderate stability of the learned weights and candidate sets. These proof-of-concept results support participant-specific composite balance evaluation for candidate selection in perturbation-based human-in-the-loop experiments.

Reasoning representations are increasingly used as explanations for large language model outputs. Yet they are typically evaluated with model-centric criteria, such as answer accuracy and faithfulness, leaving it unclear whether they help people evaluate model responses. In this work, we study reasoning representations as human-facing interfaces rather than proxies for model reasoning ability. We conduct a controlled human study of six reasoning formats across tasks of varying complexity, supported by a web-based framework that randomizes task domains, problem instances, and representation order. The study collects fine-grained judgments of structural understanding, error detection and localization, and trust calibration. Our study shows a mismatch between perceived preference and support for human evaluation. Participants prefer planning- and decomposition-based representations, but simpler chain-of-thought traces better support verification, trust, and interpretability. Preferred representations also introduce calibration risks, with more false alarms on correct traces and high trust despite low willingness to verify.

LLM-based factuality judges provide scalable evaluation signals, but their metrics are often systematically biased relative to human judgments. We study human-anchored factuality evaluation under limited annotation budgets, where judge predictions on the full dataset are combined with human labels on a small selectively sampled subset to obtain statistically valid estimates. The efficiency of this approach depends critically on which examples receive human annotation: in factuality evaluation, judge-human misalignment is not driven solely by low confidence, but also by structured failure modes such as incomplete evidence, temporal mismatch, unverifiable claims, and rubric misalignment. To exploit this structure, we introduce a factuality-specific annotation policy design pipeline that uses failure-space analysis (FSA) to derive diverse predictive signals for modeling human-judge misalignment. On an internal reference-based factuality evaluation system (AutoFA) and RAGTruth, where judge-predicted estimates substantially underestimate human-annotated factual accuracy, our FSA-guided policy improves annotation efficiency over uniform sampling and uncertainty-driven baselines, achieving effective-sample-size gains of 40.3% on AutoFA and 27.1% on RAGTruth.

Recent efforts have aimed to automate scientific diagram generation from paper content (Lin et al., 2026; Zhu et al., 2026a). However, fully satisfying an author's visual and communicative preferences in a single turn is challenging: in our formative user study (N = 14), all participants requested further revisions after viewing an initial draft, and 86% of them rated the refined diagrams as more satisfactory. Despite the clear demand, the multi-turn workflow remains largely underexplored. To bridge this gap, we present MTPaperBananaBench, a benchmark for multi-turn diagram generation containing 292 images annotated with 3,518 user requirements. To reduce expensive human studies and enable scalable benchmarking, we construct a user simulator that, at each turn, identifies unsatisfied requirements and converts k of them into natural language feedback. Evaluating both requirement satisfaction and overall diagram quality reveals two key failure modes shared across baseline multiturn systems: (1) quality drift, where diagram quality progressively declines over turns, and (2) forgetting, where previously implemented features are lost in subsequent turns. To address these issues, we introduce PaperBanana-Interact, a multi-agent system that refines diagrams via an internal critique-and-refine loop. PaperBanana-Interact consistently improves rather than degrades diagram quality across turns, outperforming baselines by 11.9-18.6 points in quality score and reducing forgetting by 3.7-6.2 points.

Evaluating research ideas generated by LLMs is difficult because their scientific value cannot be fully determined by objective criteria, and no single reference answer specifies what counts as a good idea. To address this challenge, we introduce Ideation Arena, a battle style platform that evaluates research ideas through pairwise human assessment. Ideation Arena evaluates ideas generated by 14 frontier LLMs and 5 research agent architectures built on 2 base models. To ensure a common starting point, Ideation Arena builds shared literature contexts from papers familiar to the participating researchers and provides the same contexts to all LLMs and agents. We collect over 6,000 double blind pairwise comparisons from 105 active computer science researchers and construct an Elo rating leaderboard of proposal-stage expert preferences in computer science under a shared closed-context protocol. We validate the rankings through interrater agreement and robustness analyses, showing that the leaderboard remains stable under changes in annotator composition and domain coverage. Our results show substantial variation in agent effectiveness, with some frameworks improving ideation quality over their backbones and others offering little benefit or even underperforming their base models. We further construct Ideation Arena Eval, a benchmark for assessing whether automated evaluators align with human preferences in research ideation. Experiments with current LLM judges show that they still cannot reliably reproduce expert preferences, with the best judge reaching 72.56% Soft Accuracy on Overall Quality. Our code, data, and leaderboards are available at https://github.com/foss12138/Research-Ideation-Arena.

Evaluating conversational voice agents at scale re- quires reliable assessment methods that capture both observ- able interaction quality and the contextual judgment typically provided by human evaluators. We investigate LLM-as-a-Judge evaluation by comparing human judgments with GPT-4.1 and GPT-5 on telecom and retail voice-agent conversations, across conversational quality and safety dimensions. The same interac- tions are scored under three evaluation configurations, p0, p1, and p2, to test whether automated judgments are sensitive to the evaluation setup and whether observed patterns generalize across configurations and judge models. Beyond aggregate agreement, we examine metric-level correlations, evaluator consistency, and systematic human-LLM disagreement to identify which conver- sational attributes can be judged reliably by automation and which remain sensitive to interpretation and context. Effective voice-agent evaluation is also shaped by pipeline-level factors such as speech generation, streaming, and error propagation across ASR, reasoning, and tool-calling stages, motivating our focus on comparing how human and LLM judges score the same interactions end to end. Our results show that LLM- based evaluation can serve as an effective component of large- scale voice-agent assessment, but that its reliability is metric- and configuration-dependent rather than uniform. This pro- vides an empirical framework for identifying which metrics suit automated evaluation and supports hybrid pipelines in which LLM judges handle scalable assessment while human evaluators remain engaged for metrics that demand contextual interpretation and higher-confidence judgment.

Agentic large language model (LLM) systems are reshaping scientific workflows in chemistry and drug discovery, but evaluating their open-ended, tool-augmented outputs remains a fundamental bottleneck. The LLM-as-a-Judge paradigm has emerged as a scalable alternative, but existing drug discovery benchmarks deploy LLM judges without validating their alignment with human experts. In this work, we present an LLM-as-a-Judge evaluation framework for ChatInvent, an agentic drug discovery assistant deployed at AstraZeneca, with five contributions. First, we define four output-quality evaluation dimensions---Completeness, Relevancy, Structural Clarity, and Scope Adherence---alongside deterministic Tool Call Correctness checks. Second, we validate the judge through a human alignment study with five expert annotators, comparing Gemini 3.1 Pro, Claude Opus 4.7, GPT-5, and Llama 3.1 70B as candidate judges. Third, we optimize the best-performing judge using few-shot demonstrations of human-annotated examples, improving alignment with the human majority vote from 0.80 to 0.86. Fourth, applying the optimized judge to 70 held-out questions, we surface concrete limitations and find no strong evidence that informal phrasing degrades output quality; it may, however, still be helpful to have the LLM rewrite the original question before querying the agent. Finally, we extend the framework to 38 adversarial questions that are ambiguous, invalid, out-of-scope or ethically sensitive, and show that the agent's refusal behavior is guided by the stated intent of a request. Our framework provides a reusable template for human-aligned evaluation of agentic systems in scientific domains.

LLM-as-a-Judge, which leverages a large language model to evaluate natural language generated by another AI application or model, has become a standard, scalable approach for accelerating and extending costly human evaluation. Yet most work treats a judge as a static artifact, evaluating it once at construction or against a fixed benchmark. We argue instead that an LLM judge operating in a deployed system is better understood as having a lifecycle. It must be built, trained, deployed, and continuously maintained as the surrounding data evolves, and each phase poses distinct technical and operational challenges. We present such a lifecycle for the LLM judges that evaluate recommendation explanations at Netflix. Everything we report comes out of a series of controlled online member-facing experiments, in which our pipeline generated and the judges assessed hundreds of thousands of distinct show-level explanations per week across a changing catalog. Our framework has four phases. (I) Birth defines the evaluation criteria and builds curated benchmark datasets with human labels and rationales. (II) Training refines the judges' rubrics via Reasoning-Aligned Rubric Tuning (RART), which uses a meta-judge over reasoning output as the learning signal. (III) Deployment puts one judge in two online roles, quality gating and reflective generation. (IV) Monitoring runs a continuous Human-in-the-Loop (HITL) alignment process that detects drift and triggers re-tuning behind a human review gate. We report results from a five-week online A/B test over tens of millions of members on the Netflix mobile app, in which judge-aligned explanations shifted member viewing toward novel content (previously unwatched) and increased successful browse-to-play sessions relative to a no-explanation control, with no quality-related escalations.

Humanoid motion tracking is central to teleoperation and whole-body imitation, yet evaluation often disagrees with what people perceive in videos. Kinematic errors average per-frame pose differences but miss the physical artifacts that matter most, particularly unstable support and incorrect contacts such as foot skating and mistimed touch-downs. Meanwhile, widely used test suites are small and lack the diversity needed to stress contact-rich, long-horizon behaviors. We introduce HumanTracker to make humanoid tracking evaluation both perceptually aligned and scalable. The HumanTracker benchmark contains approximately 153 hours of optical motion trajectories from multiple professional performers, organized into four motion families with text labels for fine-grained diagnosis. We further propose HumanScore, a preference-aligned metric trained on 12K motion pairs containing 24K motions. Across representative state-of-the-art trackers, HumanScore better predicts human preferences and reveals contact and stability failures that kinematic metrics often miss.

Machine learning-based Intrusion Detection Systems (IDS) have demonstrated superior performance in securing Unmanned Aerial Vehicle (UAV) networks. However, the 'black-box' nature of these models, combined with the high dimensionality of multimodal cyber-physical data, poses significant interpretability challenges. Static visualization dashboards may struggle to present complex relationships among multimodal cyber-physical features in a form that is easy for operators to inspect and interpret. To address this, we propose a Conversational XAI interface powered by Large Language Models (LLM) to facilitate on-demand investigation. In a controlled experiment with participants, we systematically evaluated the impact of this conversational interface versus a traditional XAI Dashboard on operator understanding, trust, and reliance during post-incident auditing tasks. Our results suggest that the conversational interface was perceived as more useful than the dashboard, potentially because it helped participants access and synthesize relevant information more easily. However, this benefit was accompanied by a lower level of appropriate self-reliance, indicating a potential risk of over-reliance. One possible interpretation is that the natural-language responses made the AI advice easier to accept, which may have reduced participants' tendency to verify the underlying evidence when the IDS was incorrect. These findings point to a potential trade-off in human-AI collaboration for UAV intrusion auditing: interaction mechanisms that improve perceived usability may also increase the risk of inappropriate reliance. We conclude by discussing design implications for future XAI systems that balance seamless interaction with cognitive forcing functions to foster appropriate reliance.

AI tools that help people judge online claims are usually evaluated while the tool is present. This paper asks a different question: after using such a tool, what can the user still do on their own? I call this epistemic transfer. It refers to the effect of prior AI-assisted verification on later unassisted performance on new claims. In this paper, I make three contributions. First, I distinguish epistemic transfer from nearby outcomes such as correction effects, trust, reliance, and human--AI team performance. Second, I introduce two simple quantities for studying it: the Epistemic Transfer Effect (ETE), which compares delayed unassisted performance across conditions, and Tool-Removal Cost (TRC), which measures the immediate drop in performance when the tool is taken away. Third, I turn these ideas into a practical evaluation protocol that can be used in online experiments or field studies. The protocol combines answer-first and evidence-first AI conditions with active-practice and no-practice controls, delayed tests on held-out claims, behavioral measures, and participant- and item-level analyses. Putting ETE and TRC together yields a diagnostic space that separates capability building, capability plus tool advantage, epistemic inertness or de-skilling, and verification on loan. The point is not that every AI tool must teach. The point is that when independent judgment matters, we should test not only whether a tool helps now, but also what it leaves behind.

Residential virtual power plants (VPPs) can provide grid flexibility by shifting household demand, but physical flexibility becomes dependable capacity only when residents authorize a plan and the promised response is delivered. Existing benchmarks evaluate control but omit event-specific authorization. We present EnergyBridge, a benchmark and agent framework connecting capacity reporting, household authorization, and physical execution. It combines region-specific EnergyPlus environments for Tianjin and Berlin with an LLM-based User Participation Simulator. Against 584 persona- and event-matched human role-play judgments, the LLM-based User Participation Simulator preserves method ordering with a 5.3-point mean absolute acceptance error. Across conventional controllers and agent baselines, EnergyBridge achieves the highest simulated authorization, lowest event-window energy, and the most reliable capacity commitment in both regions. We release human data and codes for reproducible human-centered grid-flexibility research: https://github.com/Agentic-Intelligence-Lab/EnergyBridge.

Youth increasingly turn to AI chatbots for social and emotional support, raising concerns about how these systems respond, especially in high-stakes situations. However, existing child safety evaluations of AI lack grounding in real-world harms that youth experience, rely on unvalidated assumptions about what counts as an appropriate output (e.g., refusal), and typically focus on detecting adversarial prompts or surface-level harms in outputs only. Thus, these evaluations can fail to detect responses that pose harm to youth in practice. To better understand the limitations of current evaluation practices, we conducted interviews with 19 practitioners working directly with youth in vulnerable situations, including social workers, therapists, and psychologists, asking them to reflect on chatbots' responses to risky situations commonly faced by youth, as established in prior empirical work. Practitioners identified chatbot behaviors likely to cause harm as well as those that could meaningfully support youth in difficult moments, discussed the role that chatbots should (and should not) play in these interactions, and offered concrete recommendations for improving chatbot responses. Based on these findings, we provide recommendations for AI child safety evaluation and infrastructure, and highlight the need for incorporating practitioners' perspectives into safety work.

While human evaluation is the gold standard in many NLP tasks, it suffers from prohibitive costs and poor scalability. When identifying top-performing models, typical evaluation protocols waste effort by exhaustively evaluating all models on the entire benchmark, a safe but inefficient approach. In this work, we formalize multi-model human evaluation as a best-arm identification problem in a multi-armed bandit setup with correlated arms, where pulling an arm corresponds to human-evaluating a model. By sampling adaptively based on the intermediate model rankings obtained on the samples so far, we can focus the annotation budget on the most competitive models. We prove the optimality of the proposed algorithms and show that it improves discrimination between top-performing models. This makes evaluations faster, cheaper and more aligned with large-scale competition evaluation goals.

Human-centered assessment tasks, which are essential for systematic decision-making, rely heavily on human judgment and typically lack verifiable ground truth. Existing approaches face a dilemma: methods using only human judgments suffer from heterogeneous expertise and inconsistent rating scales, while methods using only model-generated scores must learn from imperfect proxies or incomplete features. We propose Aggregate-then-Calibrate (AtC), a two-stage framework that combines these complementary sources. Stage-1 aggregates heterogeneous comparative judgments into a consensus ranking using a rank-aggregation model that accounts for annotator reliability. Stage-2 calibrates any predictive model's scores by an isotonic projection onto the order, enforcing ordinal consistency while preserving as much of the model's quantitative information as possible. Theoretically, we show: (1) modeling annotator heterogeneity yields strictly more efficient consensus estimation than homogeneity; (2) isotonic calibration enjoys risk bounds even when the consensus ranking is misspecified; and (3) AtC asymptotically outperforms model-only assessment. Across semi-synthetic and real-world datasets, AtC consistently improves accuracy and robustness over human-only or model-only assessments. Our results bridge judgment aggregation with model-free calibration, providing a principled recipe for human-centered assessment when ground truth is costly, scarce, or unverifiable.

Benchmark suites assess model capability on controlled tasks; large-scale conversation corpora capture naturalistic use without user feedback; and in-interface feedback mechanisms record satisfaction without task purpose. Together, they leave a critical gap in LLM evaluation: no existing infrastructure routinely links interaction trajectories to user-defined outcomes. We introduce MonitrLLM, open-source infrastructure for community-centered LLM evaluations that links full conversation transcripts to user-reported task intent and outcome assessments, treating all three as primary evaluative signals rather than optional metadata. To demonstrate the value of this approach, we conducted a two-week feasibility pilot with 26 college students using ChatGPT, collecting 206 evaluation reports with full conversation transcripts. The findings from our pilot demonstrate the value of connecting conversation trajectories with user-reported outcomes. For instance, despite reporting high average satisfaction (4.19/5) with their LLM interactions, participants also experience a substantial 23.1% failure rate on their goal tasks. We also find that multi-turn conversations are reported as failing at 2.5 times the rate of single-turn exchanges, a pattern that reframes extended interaction as a signal of difficulty rather than engagement. We conclude by discussing the value of incorporating direct user feedback with observational data for robust LLM evaluations, and the possibilities for infrastructure that enables this goal.

Evaluations of LLM-assisted qualitative coding almost universally measure model performance as agreement with human coders, a practice that presumes human coding is the standard to approximate. This study provides empirical evidence that the presumption fails in ways agreement metrics cannot detect. Five LLM systems and three trained human coders independently applied a 72-item hierarchical codebook to 2,560 educator messages from a K-12 AI platform. Beyond conventional agreement analysis, an independent domain expert judged 855 pairwise comparisons of code sets blind to source, treating human and machine sources symmetrically. The two evaluation approaches diverge in both directions. Human-LLM agreement (mean Jaccard 0.30) falls well below human-human agreement (0.52), which standard practice would read as inferior LLM coding, yet the blind verifier preferred human and LLM coding at indistinguishable rates (51.5% vs. 48.5%, p = 0.537), and a Bradley-Terry ranking placed two LLMs above two of three human coders. For several substantive codes, human consensus encoded shared bias that the verifier rejected in favor of the LLM interpretation. Agreement-based evaluation is therefore insufficient for automation decisions, and the study demonstrates a transferable verification protocol and a code-level division-of-labor framework.

Context: Generative AI (GenAI) and Large Language Models (LLMs) are increasingly used for academic tasks in software engineering and beyond, including systematic literature reviews (SLRs). However, while capable of summarizing text, there is no guarantee they can meet the rigour, reliability, and transparency that SLRs require. Objectives: To support researchers intending to conduct SLRs using GenAI or those conducting empirical studies evaluating how well GenAI supports SLR tasks. Methods: First, we conducted a rapid review to identify studies that propose guidelines for evaluating and using GenAI and LLMs to support SLRs. Second, we drew on thought experiments, relevant guidance from the literature, and our own experience conducting SLRs and evaluating tools to develop recommendations for how to use and assess GenAI in the context of SLRs. Results: We discuss the problems researchers face when evaluating GenAI for SLRs. We identify and explain process issues to consider when planning, conducting, and reporting both SLRs using GenAI and evaluations of GenAI tools. Finally, we summarize our results as a set of process recommendations, which we name GUEST (GenAI Use and Evaluation in SLR Tasks). Conclusion: We argue that GenAI requires human oversight and is not currently capable of unsupervised systematic studies. However, it offers the prospect of cost-effective assistance for some repetitive tasks and for additional validation of some complex tasks. Our GUEST recommendations should help software engineering researchers both to conduct and report trustworthy SLRs using GenAI and to provide rigorous independent evaluation studies.