Abstract As climate-induced extreme weather events intensify, aging homes in low-income and rural communities face mounting energy burdens and vulnerabilities. Emerging digital twin technologies, combined with unmanned aerial vehicles, infrared thermography, computer vision, and artificial intelligence, offer promise for efficient building energy auditing—an essential step for government programs to release funds. However, the adoption of such technologies by certified auditors remains limited due to systemic barriers. To address the gap, this research starts with a systematic literature review to identify key factors influencing the implementation of these advanced technologies, considering both technical and managerial constraints. Afterward, a framework derived from the literature is used to support focus groups and surveys with statewide weatherization audit agencies to capture local perceptions, concerns, and needs. The descriptive results suggest that, despite engineers focusing on pushing the technological boundaries, potential end users tend to prioritize managerial aspects such as cost, training, ease of use, addressing outstanding needs, compliance, and data security. The results also illustrate the silhouette of future weatherization audit scenarios, outlining core required features. The findings inform the participatory design of a digital twin platform intended to democratize access to energy auditing and enable data-driven retrofit decisions within publicly administered weatherization assistance programs.
Child sexual abuse (CSA) is a major global public health, human rights, and social problem with adverse consequences on the health and well-being of children. This study examined the experiences of Ghanaian female survivors of CSA regarding the perpetrators of CSA, the reporting of CSA incidents, and their emotions/feelings after they were sexually abused. The study used secondary qualitative data from a larger study by the Ministry of Gender, Children, and Social Protection in 2018, which involved fifteen female survivors of CSA. The transcripts of the fifteen participants were analyzed using thematic analysis. The findings revealed that perpetrators of CSA were persons closer to survivors, such as boyfriends, neighbors, siblings’ friends, relatives, and class teachers. Additionally, while most participants reported their CSA incidents, others were unable to disclose those acts. Some survivors of CSA felt distressed when they saw perpetrators, while others felt anger, fear, and mistrust of men after their abuse. This study advocates for the development of programs that empower families to understand, prevent, and respond to cases of CSA. Also, there is a need to prioritize mental health support for CSA survivors.
The Trauma-Informed Programs and Practices for Schools (TIPPS) framework is a system-level model that promotes safe, inclusive, and supportive school environments. A key feature of TIPPS is implementation of a comprehensive survey for needs assessment and progress monitoring organized according to the model’s 10 core pillars. This study conducts an initial psychometric evaluation of the 50-item TIPPS survey, investigating its factor structure, reliability, and convergent validity with scales assessing related aspects of school climate. Participants were 300 K-12 school personnel recruited through an online panel to complete a one-time questionnaire including the TIPPS survey and other measures. Internal structure was examined using confirmatory factor analysis (CFA). Reliability was assessed using standard procedures. Convergent validity was examined via CFA in relation to four established school climate measures. A CFA of the hypothesized 10-pillar structure of the TIPPS survey displayed good model fit. Two of the 10 pillars were represented as manifest variables. Standardized factor loadings for the eight pillars represented as latent variables were statistically significant, and the majority were greater than 0.50. Seven out of the eight pillars for which Cronbach’s alpha was appropriate demonstrated high reliability (above 0.70). Correlations among the pillars were positive and statistically significant. Moreover, each TIPPS pillar had positive and statistically significant associations with each of the four latent school climate factors. Findings suggest the TIPPS survey corresponds with the hypothesized 10-pillar structure and exhibits acceptable validity and reliability, supporting its value as a tool for needs assessment and progress monitoring.
Recent advances in large language models (LLMs) have enabled agentic systems that translate natural language intent into executable scientific visualization (SciVis) tasks. Despite rapid progress, the community lacks a principled and reproducible benchmark for evaluating these emerging SciVis agents in realistic, multi-step analysis settings. We present SciVisAgentBench, a comprehensive and extensible benchmark for evaluating scientific data analysis and visualization agents. Our benchmark is grounded in a structured taxonomy spanning four dimensions: application domain, data type, complexity level, and visualization operation. It currently comprises 108 expert-crafted cases covering diverse SciVis scenarios. To enable reliable assessment, we introduce a multimodal outcome-centric evaluation pipeline that combines LLM-based judging with deterministic evaluators, including image-based metrics, code checkers, rule-based verifiers, and case-specific evaluators. We also conduct a validity study with 12 SciVis experts to examine the agreement between human and LLM judges. Using this framework, we evaluate representative SciVis agents and general-purpose coding agents to establish initial baselines and reveal capability gaps. SciVisAgentBench is designed as a living benchmark to support systematic comparison, diagnose failure modes, and drive progress in agentic SciVis. The benchmark is available at https://scivisagentbench.github.io/.
. Iterative decoder failures of quantum low density parity check (QLDPC) codes are attributed to substructures in the code's graph, known as trapping sets, as well as degenerate errors that can arise in quantum codes. Failure inducing sets are subsets of codeword coordinates that, when initially in error, lead to decoding failure in a trapping set. The purpose of this paper is to examine failure inducing sets of QLDPC codes under syndrome-based iterative decoding. As for classical LDPC codes, we show that absorbing sets play a central role in understanding decoder failures. Raveendran and Vasic [11] initiated the study of quantum trapping sets, where beyond the classical-type trapping sets, they identified rigid symmetric structures (a.k.a. symmetric stabilizers) responsible for degenerate errors. In this paper, we show that this behavior is part of a much more general phenomenon that can be described by the absorbing set framework.
