Santa Clara University

Interactive artificial intelligence (AI) is rapidly reshaping human-centered robotics by moving beyond algorithmic efficiency toward real-time adaptability, transparency, and design for people and contexts. Building on the inaugural InterAI Workshop at IEEE RO-MAN 2024, this second edi- tion, proposed for HRI 2026, will convene the robotics and HRI communities to examine how interactive AI can en- able robust, trustworthy systems that operate seamlessly in dynamic, real-world environments. The half-day hybrid program will feature two keynote talks, and a peer-reviewed paper oral presentation session. Distinct from prior events, the workshop spotlights the integration of generative and embodied AI with human-in-the-loop learning and real-time decision making, as well as methods for explainability, eval- uation, and safety. The workshop aims to catalyze a collabo- rative agenda that bridges interactive AI technologies and human-centered robotic systems.

We investigate what structure emerges in 3D Gaussian Splatting (3DGS) solutions from standard multi-view optimization. We term these Rendering-Optimal References (RORs) and analyze their statistical properties, revealing stable patterns-log-normal sca

DNA storage has matured from concept to practical stage, yet its integration with neural compression pipelines remains inefficient. Early DNA encoders applied redundancy-heavy constraint layers atop raw binary data-workable but primitive. Recent neural codecs compress data into learned latent representations with rich statistical structure, yet still convert these latents to DNA via naive binary-to-quaternary transcoding, discarding the entropy model's optimization. This mismatch undermines compression efficiency and complicates the encoding stack. A plug-in module that collapses latent compression and DNA encoding into a single step. SCONE performs quaternary arithmetic coding directly on the latent space in DNA bases. Its Constraint-Aware Adaptive Coding module dynamically steers the entropy encoder's learned probability distribution to enforce biochemical constraints-GC balance and homopolymer suppression-deterministically during encoding, eliminating post-hoc correction. The design preserves full reversibility and exploits the hyperprior model's learned priors without modification. Experiments show SCONE achieves near-perfect constraint satisfaction with negligible computational overhead (< 2% latency), establishing a latent-agnostic interface for end-to-end DNA-compatible learned codecs.

Potassium competitive acid blockers (PCABs) are superior to proton pump inhibitors (PPIs) for healing of Los Angeles (LA) class C/D erosive esophagitis and are approved for nonerosive gastroesophageal reflux disease (GERD) given their efficacy measured by heartburn-free days. However, they are more expensive than PPIs. We estimated the cost-effectiveness of PCABs compared with PPIs for the management of GERD. A decision tree was constructed for the base case of a patient with GERD. We tested scenarios in which nonerosive reflux disease, LA A/B, or LA C/D esophagitis was present, comparing PCABs with PPIs as first-line therapy, including step up therapy and/or class switching if symptoms persisted. Using publicly available cost estimates, quality-adjusted life years (QALYs) were compared using a willingness-to-pay threshold of $100,000/QALY from a societal perspective at 6 months. The cost of PCABs for GERD was $3,240 more than PPIs, with an incremental cost-effectiveness ratio of $144,208/QALY. When evaluating GERD phenotypes separately, the incremental cost-effectiveness ratio was consistently over our willingness-to-pay threshold. One-way analyses showed the most influential parameters were PCAB cost and efficacy for heartburn-free days. Probabilistic sensitivity analysis favored PPIs 71.7% out of 100,000 iterations. Reducing one-week costs to below $91 would make PCABs a cost effective first-line strategy across all GERD phenotypes. Despite PCAB efficacy, PPIs are a cost-effective strategy for GERD treatment, and can be positioned ahead of PCAB escalation on this basis. Reducing PCAB costs or accepting a higher willingness-to-pay threshold would influence this finding, although PCABs may be favored in some current situations as evidenced by probabilistic sensitivity results.

As both public and Catholic schools across the U.S. have become increasingly diverse in their student bodies, scholars have called for K-12 teachers to abandon deficit discourses, which blame students of colour and students from low-income households for discrepancies in their learning outcomes as compared to those of white and middle-class students. Instead, educators are urged to adopt an asset orientation toward students, especially those from minoritized communities. The purpose of this manuscript is to explicate the dispositions that comprise an asset orientation; examine the secular theories of education from which they arise; and explore the extent to which these pieces align with a Catholic view of schooling. To illustrate what an asset orientation might look like, the paper also provides examples of asset-oriented reflections from Catholic school teachers who learned about and intentionally practiced these dispositions. The paper concludes by proposing a Catholic concept of an asset orientation.

The genusStreptococcuscontains some of the most important commensals and pathogens of the human microbiome. To obtain the fatty acids required for cell membranes,Streptococcuseither produce fatty acidsde novothrough the fatty acid biosynthesis (fab) pathway or uptake host fatty acids through the fatty acid kinase (fak) pathway. Although both thefabandfakpathways represent potential therapeutic targets to prevent or treat infection, progress is limited because of an incomplete understanding of taxon-to-taxon variability in streptococcal lipid metabolism. Here, we examined the role ofde novomonounsaturated fatty acid (MUFA) synthesis in physiology and virulence-associated traits inStreptococcus mutans, Streptococcus pyogenes,andStreptococcus pneumoniae,three major pathogens that cause disease at distinct body sites. In all three species, deletion offabMabolished MUFA production and caused severe growth defects, decreased stress tolerance, increased antibiotic susceptibility, and defects in cell viability, morphology, and division. InS. mutans,loss offabMalso markedly reduced competence signaling and production of the mutacin IV bacteriocin. Deletion offabMincreased susceptibility to killing by human neutrophils inS. mutansandS. pneumoniae,but notS. pyogenes. Together, these findings illustrate that MUFA synthesis is broadly important for streptococcal physiology and cell membrane homeostasis, while its contribution to pathogenesis is strongly species- and context-dependent, providing leads to guide development of novel therapeutic and/or preventative strategies.

Work-related musculoskeletal disorders (WMSDs) persist in material-handling jobs where lifting, twisting, and carrying induce high, localized muscle demands. This paper presents a sensor-driven framework that (i) detects biomechanical strain from surface electromyography (sEMG) and (ii) quantifies the impact of a passive back-support exoskeleton during industrially relevant tasks. With data from 20 participants performing standardized tasks with and without the device, we introduce a data-driven strain labeling method that replaces ad-hoc thresholds with piecewise linear regression to identify individualized strain onset. A compact deep neural network handles severe class imbalance via SMOTE and decision-threshold optimization, yielding 83.5% overall accuracy and a macro-averaged F1-score of 0.70 for binary strain classification. Muscle-specific analyses reveal significant reductions in biceps and oblique activation (p < 0.001) alongside compensatory increases in erector spinae and lower-limb activity, indicating load redistribution rather than uniform offloading. The result is a scalable, real-time approach that captures both when strain begins and how effort shifts across muscle groups, capabilities that traditional peak-sEMG or subjective assessments miss. By uniting wearable sensing, automated strain onset detection, and imbalance-aware learning, this work advances objective, continuous, and human-centered ergonomic monitoring and provides actionable evidence for the deployment of passive exoskeletons in smart industrial environments.