Generative artificial intelligence (GenAI) systems are increasingly used by organizations to deliver information to consumers, patients, students, employees, and citizens. These systems can hallucinate, producing plausible but inaccurate responses. A central question for AI-advised decisions is therefore not only whether users rely on inaccurate information, but whether they recognize that a response may require verification. To answer this question, we review emerging empirical evidence relevant to hallucination detection in goal-directed interactions, with a focus on organization-backed AI advisors. We distinguish three constructs that existing studies often conflate: whether users are skeptical of information presented, whether they verify it (distinguishing attempted from successful verification), and whether the result of verification affects reliance on the information. Across studies examining product search, medical decision-making, content generation, and chatbot-assisted tasks, several patterns emerge. Nearly all studies measure reliance, while variables such as user skepticism and verification of the information are more often targeted by an intervention than measured directly. The cues used to prompt scrutiny of the AI response are predominantly related to the AI output, such as source citations. Whereas general warnings about the risk of hallucinations are among the most deployable interventions for organizations, they show the weakest and most mixed effects in the studies reviewed. More targeted cues about where AI systems tend to fail appear promising, however evidence remains limited. Although the existing literature posits that users may be more likely to scrutinize responses related to particular areas of content, no studies varied the content category, leaving this question open for further research. In future research, measuring skepticism and verification separately from reliance may clarify what current evidence shows, what it only implies, and which questions require further exploration.
We prove that an affine hypersurface (called monolinear hypersurface) (XY)-Y-d + h = 0, where h is an element of k[X, Z(1), ... , Z(n)] and n, d > 2 (no restrictions on the field k) is an affine hyperplane only if for any assignment of weights to Z(1), ... , Z(n), the generic translate of the weighted degree form of g := h(0, Z(1), ... , Z(n)) is separably uniruled. In particular, an affine hypersurface f = 0 defined by f is an element of k[Z(1), ... , Z(n)] is a coordinate hyperplane only if the generic translate of each of the weighted degree forms of f is separably uniruled. Separable uniruled-ness of generic translates is a rather restrictive condition on hypersurfaces. Our main theorem furnishes the first equational test of nonhyperplanarity of monolinear hypersurfaces in dimensions > 4. Non-hyperplanarity of the Russell 3-fold and the Asanuma 3-folds (in positive characteristic), which have featured prominently in the context of linearization of torus actions and Zariski-cancellation, readily follows from our main theorem.
Semiconductor materials play a central role in current and future electronics technologies. From microprocessors and advanced computers to optical components, device functionality is dependent on the creation and control of point defects in semiconductors. Incorporating a fundamental understanding of defect kinetics, including formation, migration, and chemistry, is essential for advancing materials science, assessing their device impact, ensuring the reliability of modern electronics, and leveraging new materials for next-generation technologies. This article explores the kinetics of point defects from experimental observations and atomistic modeling, to dynamical multiscale descriptions of defect kinetics. A survey of experiments reveals the important role of kinetics in defect behavior during synthesis, implantation doping, radiation exposure, and long-term defect evolution, while highlighting the impact of evolving material properties on device performance. Atomistic modeling, including molecular dynamics and density functional theory, is surveyed emphasizing its ability to describe dynamical behavior and predict kinetic pathways that govern defect evolution in semiconductors. Dynamical and multiscale modeling methods that integrate experimental and atomistic defect properties into continuum-scale codes are examined for their role in bridging atomic-scale defect behavior to device-level performance. By addressing critical challenges and revealing the inherent difficulties in modeling and experimental validation, this article aims to advance the understanding of defect kinetics and provide insights into the short-term and long-term reliability of materials and devices.
The NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Vaginal Cancer outline the recommended diagnostic workup, staging considerations, and treatment options for this rare malignancy. Because vaginal cancer is uncommon and shares many biologic and clinical characteristics with cervical cancer, several management recommendations, particularly systemic therapy, are extrapolated from evidence and practices established for cervical cancer. This guideline excerpt summarizes key components of management, including diagnostic evaluation and workup, principles of staging, pathology considerations, radiation therapy principles, and primary treatment recommendations for both early-stage and advanced disease. It also details approaches to manage relapses, including locoregional recurrence and distant metastases, and provides an in-depth overview of systemic therapy recommendations for vaginal cancer. J Natl Compr Canc Netw 2026;24(3):101-126 doi:10.6004/jnccn.2026.0011
Retailers struggle with late deliveries, thus motivating research to improve e-fulfillment performance. Studies have primarily investigated order processing and delivery individually but have ignored the interplay between these two e-fulfillment activities. The Theory of Swift and Even Flow (TSEF) provides a useful frame for examining the e-fulfillment process, yet it neglects important behavioral factors. We elaborate the TSEF using logic from the Queue-Length Visibility and Misperception of Feedback Dynamics perspectives to unveil how behaviors in order processing and delivery contribute to delivery performance. We analyze 11,241 orders from a major Vietnamese retailer using econometric methods informed by practitioner interviews. We find a concave relationship between order processing time (OPT) ratio (defined as the proportion of planned lead time consumed by order processing) and lateness. Late orders have OPT ratios exceeding 25% of the planned lead time, and they exhibit higher OPT variance. We also find a U-shaped relationship between OPT ratio and order delivery time (ODT); expediting deliveries mitigates delays until OPT ratios reach a threshold of 58%. Finally, we argue that workers and managers prioritize the processing of focused orders. Understanding behaviors in the e-fulfillment process offers insights that extend the TSEF, new research areas, and managerial implications.
