❌

Normal view

Received — 15 December 2025 ⏭ Favourites

Benchmarking AI Models in Software Engineering: A Review, Search Tool, and Unified Approach for Elevating Benchmark Quality

✇cs.AI, q-bio.NC updates on arXiv.org
By: Roham Koohestani, Philippe de Bekker, Beg\"um Ko\c{c}, Maliheh Izadi
15 December 2025 at 13:00
arXiv:2503.05860v3 Announce Type: replace-cross Abstract: Benchmarks are essential for unified evaluation and reproducibility. The rapid rise of Artificial Intelligence for Software Engineering (AI4SE) has produced numerous benchmarks for tasks such as code generation and bug repair. However, this proliferation has led to major challenges: (1) fragmented knowledge across tasks, (2) difficulty in selecting contextually relevant benchmarks, (3) lack of standardization in benchmark creation, and (4) flaws that limit utility. Addressing these requires a dual approach: systematically mapping existing benchmarks for informed selection and defining unified guidelines for robust, adaptable benchmark development. We conduct a review of 247 studies, identifying 273 AI4SE benchmarks since 2014. We categorize them, analyze limitations, and expose gaps in current practices. Building on these insights, we introduce BenchScout, an extensible semantic search tool for locating suitable benchmarks. BenchScout employs automated clustering with contextual embeddings of benchmark-related studies, followed by dimensionality reduction. In a user study with 22 participants, BenchScout achieved usability, effectiveness, and intuitiveness scores of 4.5, 4.0, and 4.1 out of 5. To improve benchmarking standards, we propose BenchFrame, a unified framework for enhancing benchmark quality. Applying BenchFrame to HumanEval yielded HumanEvalNext, featuring corrected errors, improved language conversion, higher test coverage, and greater difficulty. Evaluating 10 state-of-the-art code models on HumanEval, HumanEvalPlus, and HumanEvalNext revealed average pass-at-1 drops of 31.22% and 19.94%, respectively, underscoring the need for continuous benchmark refinement. We further examine BenchFrame's scalability through an agentic pipeline and confirm its generalizability on the MBPP dataset. All review data, user study materials, and enhanced benchmarks are publicly released.

Understanding Prompt Management in GitHub Repositories: A Call for Best Practices

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hao Li · Hicham Masri · Filipe R. Cogo · Abdul Ali Bangash · Bram Adams · Ahmed E. Hassan
15 December 2025 at 13:00
arXiv:2509.12421v2 Announce Type: replace-cross Abstract: The rapid adoption of foundation models (e.g., large language models) has given rise to promptware, i.e., software built using natural language prompts. Effective management of prompts, such as organization and quality assurance, is essential yet challenging. In this study, we perform an empirical analysis of 24,800 open-source prompts from 92 GitHub repositories to investigate prompt management practices and quality attributes. Our findings reveal critical challenges such as considerable inconsistencies in prompt formatting, substantial internal and external prompt duplication, and frequent readability and spelling issues. Based on these findings, we provide actionable recommendations for developers to enhance the usability and maintainability of open-source prompts within the rapidly evolving promptware ecosystem.

MetaVoxel: Joint Diffusion Modeling of Imaging and Clinical Metadata

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yihao Liu · Chenyu Gao · Lianrui Zuo · Michael E. Kim · Brian D. Boyd · Lisa L. Barnes · Walter A. Kukull · Lori L. Beason-Held · Susan M. Resnick · Timothy J. Hohman · Warren D. Taylor · Bennett A. Landman
15 December 2025 at 13:00
arXiv:2512.10041v2 Announce Type: replace-cross Abstract: Modern deep learning methods have achieved impressive results across tasks from disease classification, estimating continuous biomarkers, to generating realistic medical images. Most of these approaches are trained to model conditional distributions defined by a specific predictive direction with a specific set of input variables. We introduce MetaVoxel, a generative joint diffusion modeling framework that models the joint distribution over imaging data and clinical metadata by learning a single diffusion process spanning all variables. By capturing the joint distribution, MetaVoxel unifies tasks that traditionally require separate conditional models and supports flexible zero-shot inference using arbitrary subsets of inputs without task-specific retraining. Using more than 10,000 T1-weighted MRI scans paired with clinical metadata from nine datasets, we show that a single MetaVoxel model can perform image generation, age estimation, and sex prediction, achieving performance comparable to established task-specific baselines. Additional experiments highlight its capabilities for flexible inference. Together, these findings demonstrate that joint multimodal diffusion offers a promising direction for unifying medical AI models and enabling broader clinical applicability.

High-Throughput Dissection of Inter-Organ Genetic Networks: A Multi-Omic Systems Biology Approach

✇Omics In Lung
By: Rana Alabdan · Hechmi Shili · Ghada Moh Samir Elhessewi · Mukhtar Ghaleb · Eman M Alanazi · Nouf Helal Alharbi · Rowida Mohammed Alharbi · Asma A Alhashmi
13 December 2025 at 19:00

SLAS Technol. 2025 Dec 11:100376. doi: 10.1016/j.slast.2025.100376. Online ahead of print.

ABSTRACT

The existing multi-omic analyses are frequently confined to individual tissues, and the regulatory picture of the systemic regulator of complex physiology and disease is hidden. To fill this gap, we have created a unified systems biology model of the high-throughput dissection of inter-organ genetic networks. Our model incorporates transcriptomic, epigenomic and proteomic analysis of five major organs (liver, kidney, heart, lung, brain) using the Multi-Omics Factor Analysis (MOFA+) tool, specifically, cross-tissue coordination. We characterized 27 evidence-heavy cross-tissue modules (FDR < 0.05) that are major hubs such as *HNF4Aenda NRF2cheng8loadmasterregulatingconstitutionembryonicstemcellularinfoncogenes recognize them. One notable observation was liver-kidney metabolic axis, significant cross-talks in hepatocyte organoids are confirmed with CRISPR knockdown, which suppresses the expression of transporters expressed by the kidney. Our work offers a scalable validated framework that goes beyond organ-centric perspectives, which can be used as a potent tool of systemic disease modelling and precision medicine.

PMID:41389879 | DOI:10.1016/j.slast.2025.100376

Received — 14 December 2025 ⏭ Favourites

Mapping the inflammatory origins of lung cancer

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Nobunari Sasaki · Mizuki Homme · Shunsuke Kitajima
12 December 2025 at 19:00

Cancer Cell. 2025 Dec 11:S1535-6108(25)00498-2. doi: 10.1016/j.ccell.2025.11.005. Online ahead of print.

ABSTRACT

How early precursor cells and their surrounding microenvironment cooperate to drive oncogenic progression in lung adenocarcinoma (LUAD) remains elusive. In this issue of Cancer Cell, Peng et al. conducted multimodal spatial-omics to comprehensively profile precancerous lung and LUAD tissues, uncovering alveolar progenitors and proinflammatory niches that co-evolve during cancer progression.

PMID:41386222 | DOI:10.1016/j.ccell.2025.11.005

Received — 12 December 2025 ⏭ Favourites

Minimal Residual Disease Detection: Bridging Molecular and Clinical Strategies for Recurrence Prevention in Gynecologic Cancers

✇MRD
By: Andi Darma Putra · Naufal Syafiq Darmawan · Aldi Tamara Rahman · Lasmini Syariatin
11 December 2025 at 19:00

Int J Mol Sci. 2025 Dec 3;26(23):11708. doi: 10.3390/ijms262311708.

ABSTRACT

Gynecologic cancers remain a major global health burden, particularly in low- and middle-income countries, with high incidence and mortality rates around 45-50%. The detection of minimal residual disease (MRD) is transforming the management of recurrence risk in gynecologic cancers through highly sensitive molecular technologies. MRD encompasses small populations of residual cancer cells or post-treatment molecular traces but remain undetectable by conventional methods. Its detection relies on circulating tumor DNA (ctDNA), circulating tumor cells (CTCs), and advanced next-generation sequencing (NGS), with ctDNA-based MRD assays having sensitivity levels between 85% and over 99%. Other technologies, such as liquid biopsies and digital PCR, are also in development. MRD status has demonstrated high predictors of recurrence and survival with positive MRD strongly associated with poor outcomes and negative MRD indicates sustained remission. However, MRD detection faces significant limitations, such as tumor heterogeneity, inconstant ctDNA levels, technical issues of false-negative results, and limited clinical accessibility. Therefore, this review presents current evidence regarding the molecular detection of MRD in gynecologic malignancies and assesses its prognostic and predictive relevance. Ultimately, MRD continuous integration into clinical practice offers a promising modality to enable early relapse detection, more precise therapeutic decision-making, and the improvement of personalized medicine access to gynecologic cancers worldwide.

PMID:41373852 | PMC:PMC12692091 | DOI:10.3390/ijms262311708

Macrophage-targeted immunocytokine leverages myeloid, T, and NK cell synergy for cancer immunotherapy

✇Cell
By: Michelle von Locquenghien · Pascale Zwicky · Ken Xie · Diego Adhemar Jaitin · Fadi Sheban · Adam Yalin · Florian Uhlitz · Chamutal Gur · Reut Sharet Eshed · Eyal David · Kfir Mazuz · Caroline Jennings Marin · Ankita Sankar · Devin Mediratta · Roberto Avellino · Assaf Weiner · Ido Amit
19 November 2025 at 08:00
MiTEs are myeloid-targeted immunocytokine prodrugs that block TREM2+ tumor-associated macrophages while activating cytotoxic lymphocytes via TME-specific IL-2 activity, eliciting strong anti-tumor efficacy in preclinical models with minimal systemic toxicity.

Pancreatic Cancer Organoids: Modeling Disease and Guiding Therapy

✇(Multiomics OR Omics) AND (Pancreatic)
By: Franck Morceau · Victoria El-Khoury · Kyeong Lee · Marc Jean Berna · Yong-Jun Kwon
11 December 2025 at 19:00

Cancers (Basel). 2025 Nov 30;17(23):3850. doi: 10.3390/cancers17233850.

ABSTRACT

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies. An unmet need exists for reliable biomarkers and in vitro models capable of predicting patient drug response to advance personalized medicine. Traditional models fail to represent the tumor's complexity and the role of the stromal environment in chemoresistance. Patient-derived organoids (PDOs) overcome these limitations, enabling multi-omics profiling and reliable drug testing for functional precision medicine. This review provides a comprehensive overview of PDAC PDO research, emphasizing the following major areas: (i) the genetic and phenotypic fidelity of PDOs, (ii) their predictive value for drug response and chemoresistance, (iii) the integration of the extracellular matrix and tumor microenvironment (TME) components, and (iv) emerging technologies. Studies confirm that PDOs faithfully represent the primary tumor's specific genetic features and retain intratumoral heterogeneity. PDO-based platforms have demonstrated a strong correlation between in vitro drug sensitivity and in vivo efficacy in xenograft models, validating their utility for identifying drug candidates, repurposing existing drugs, and determining effective combinations. Efforts are ongoing to integrate crucial TME components, like cancer-associated fibroblasts, using innovative co-culture platforms such as fused PDOs and InterOMaX, to better model desmoplasia and chemoresistance mechanisms. Furthermore, PDO technology is converging with microphysiological systems and artificial intelligence tools to facilitate high-throughput drug screening and dynamic, real-time monitoring of therapeutic effects. The integration of PDOs into biobanks and advanced screening platforms holds the potential to accelerate drug discovery and improve therapeutic outcomes for PDAC patients, if challenges related to protocol standardization and regulatory acceptance are addressed.

PMID:41375051 | PMC:PMC12690986 | DOI:10.3390/cancers17233850

AI-driven virtual cell models in preclinical research: technical pathways, validation mechanisms, and clinical translation potential

✇npj Digital Medicine
By: Chunyu Ma · Han Zhang · Yiwei Rao · Xinyu Jiang · Boheng Liu · Zhikang Sun · Zhenyu Song · Yuan Gao · Yuhao Cui · Xinyu Liu · Zedong Li
11 December 2025 at 08:00

npj Digital Medicine, Published online: 11 December 2025; doi:10.1038/s41746-025-02198-6

AI-driven virtual cell models in preclinical research: technical pathways, validation mechanisms, and clinical translation potential
Received — 11 December 2025 ⏭ Favourites

Toward an AI Reasoning-Enabled System for Patient-Clinical Trial Matching

✇cs.AI, q-bio.NC updates on arXiv.org
By: Caroline N. Leach · Mitchell A. Klusty · Samuel E. Armstrong · Justine C. Pickarski · Kristen L. Hankins · Emily B. Collier · Maya Shah · Aaron D. Mullen · V. K. Cody Bumgardner
10 December 2025 at 13:00
arXiv:2512.08026v1 Announce Type: new Abstract: Screening patients for clinical trial eligibility remains a manual, time-consuming, and resource-intensive process. We present a secure, scalable proof-of-concept system for Artificial Intelligence (AI)-augmented patient-trial matching that addresses key implementation challenges: integrating heterogeneous electronic health record (EHR) data, facilitating expert review, and maintaining rigorous security standards. Leveraging open-source, reasoning-enabled large language models (LLMs), the system moves beyond binary classification to generate structured eligibility assessments with interpretable reasoning chains that support human-in-the-loop review. This decision support tool represents eligibility as a dynamic state rather than a fixed determination, identifying matches when available and offering actionable recommendations that could render a patient eligible in the future. The system aims to reduce coordinator burden, intelligently broaden the set of trials considered for each patient and guarantee comprehensive auditability of all AI-generated outputs.

Principles2Plan: LLM-Guided System for Operationalising Ethical Principles into Plans

✇cs.AI, q-bio.NC updates on arXiv.org
By: Tammy Zhong · Yang Song · Maurice Pagnucco
10 December 2025 at 13:00
arXiv:2512.08536v1 Announce Type: new Abstract: Ethical awareness is critical for robots operating in human environments, yet existing automated planning tools provide little support. Manually specifying ethical rules is labour-intensive and highly context-specific. We present Principles2Plan, an interactive research prototype demonstrating how a human and a Large Language Model (LLM) can collaborate to produce context-sensitive ethical rules and guide automated planning. A domain expert provides the planning domain, problem details, and relevant high-level principles such as beneficence and privacy. The system generates operationalisable ethical rules consistent with these principles, which the user can review, prioritise, and supply to a planner to produce ethically-informed plans. To our knowledge, no prior system supports users in generating principle-grounded rules for classical planning contexts. Principles2Plan showcases the potential of human-LLM collaboration for making ethical automated planning more practical and feasible.

Biothreat Benchmark Generation Framework for Evaluating Frontier AI Models I: The Task-Query Architecture

✇cs.AI, q-bio.NC updates on arXiv.org
By: Gary Ackerman · Brandon Behlendorf · Zachary Kallenborn · Sheriff Almakki · Doug Clifford · Jenna LaTourette · Hayley Peterson · Noah Sheinbaum · Olivia Shoemaker · Anna Wetzel
10 December 2025 at 13:00
arXiv:2512.08130v1 Announce Type: cross Abstract: Both model developers and policymakers seek to quantify and mitigate the risk of rapidly-evolving frontier artificial intelligence (AI) models, especially large language models (LLMs), to facilitate bioterrorism or access to biological weapons. An important element of such efforts is the development of model benchmarks that can assess the biosecurity risk posed by a particular model. This paper describes the first component of a novel Biothreat Benchmark Generation (BBG) Framework. The BBG approach is designed to help model developers and evaluators reliably measure and assess the biosecurity risk uplift and general harm potential of existing and future AI models, while accounting for key aspects of the threat itself that are often overlooked in other benchmarking efforts, including different actor capability levels, and operational (in addition to purely technical) risk factors. As a pilot, the BBG is first being developed to address bacterial biological threats only. The BBG is built upon a hierarchical structure of biothreat categories, elements and tasks, which then serves as the basis for the development of task-aligned queries. This paper outlines the development of this biothreat task-query architecture, which we have named the Bacterial Biothreat Schema, while future papers will describe follow-on efforts to turn queries into model prompts, as well as how the resulting benchmarks can be implemented for model evaluation. Overall, the BBG Framework, including the Bacterial Biothreat Schema, seeks to offer a robust, re-usable structure for evaluating bacterial biological risks arising from LLMs across multiple levels of aggregation, which captures the full scope of technical and operational requirements for biological adversaries, and which accounts for a wide spectrum of biological adversary capabilities.

A Practical Framework for Evaluating Medical AI Security: Reproducible Assessment of Jailbreaking and Privacy Vulnerabilities Across Clinical Specialties

✇cs.AI, q-bio.NC updates on arXiv.org
By: Jinghao Wang · Ping Zhang · Carter Yagemann
10 December 2025 at 13:00
arXiv:2512.08185v1 Announce Type: cross Abstract: Medical Large Language Models (LLMs) are increasingly deployed for clinical decision support across diverse specialties, yet systematic evaluation of their robustness to adversarial misuse and privacy leakage remains inaccessible to most researchers. Existing security benchmarks require GPU clusters, commercial API access, or protected health data -- barriers that limit community participation in this critical research area. We propose a practical, fully reproducible framework for evaluating medical AI security under realistic resource constraints. Our framework design covers multiple medical specialties stratified by clinical risk -- from high-risk domains such as emergency medicine and psychiatry to general practice -- addressing jailbreaking attacks (role-playing, authority impersonation, multi-turn manipulation) and privacy extraction attacks. All evaluation utilizes synthetic patient records requiring no IRB approval. The framework is designed to run entirely on consumer CPU hardware using freely available models, eliminating cost barriers. We present the framework specification including threat models, data generation methodology, evaluation protocols, and scoring rubrics. This proposal establishes a foundation for comparative security assessment of medical-specialist models and defense mechanisms, advancing the broader goal of ensuring safe and trustworthy medical AI systems.

ClinicalTrialsHub: Bridging Registries and Literature for Comprehensive Clinical Trial Access

✇cs.AI, q-bio.NC updates on arXiv.org
By: Jiwoo Park · Ruoqi Liu · Avani Jagdale · Andrew Srisuwananukorn · Jing Zhao · Lang Li · Ping Zhang · Sachin Kumar
10 December 2025 at 13:00
arXiv:2512.08193v1 Announce Type: cross Abstract: We present ClinicalTrialsHub, an interactive search-focused platform that consolidates all data from ClinicalTrials.gov and augments it by automatically extracting and structuring trial-relevant information from PubMed research articles. Our system effectively increases access to structured clinical trial data by 83.8% compared to relying on ClinicalTrials.gov alone, with potential to make access easier for patients, clinicians, researchers, and policymakers, advancing evidence-based medicine. ClinicalTrialsHub uses large language models such as GPT-5.1 and Gemini-3-Pro to enhance accessibility. The platform automatically parses full-text research articles to extract structured trial information, translates user queries into structured database searches, and provides an attributed question-answering system that generates evidence-grounded answers linked to specific source sentences. We demonstrate its utility through a user study involving clinicians, clinical researchers, and PhD students of pharmaceutical sciences and nursing, and a systematic automatic evaluation of its information extraction and question answering capabilities.

Are generative AI text annotations systematically biased?

✇cs.AI, q-bio.NC updates on arXiv.org
By: Sjoerd B. Stolwijk · Mark Boukes · Damian Trilling
10 December 2025 at 13:00
arXiv:2512.08404v1 Announce Type: cross Abstract: This paper investigates bias in GLLM annotations by conceptually replicating manual annotations of Boukes (2024). Using various GLLMs (Llama3.1:8b, Llama3.3:70b, GPT4o, Qwen2.5:72b) in combination with five different prompts for five concepts (political content, interactivity, rationality, incivility, and ideology). We find GLLMs perform adequate in terms of F1 scores, but differ from manual annotations in terms of prevalence, yield substantively different downstream results, and display systematic bias in that they overlap more with each other than with manual annotations. Differences in F1 scores fail to account for the degree of bias.

Biothreat Benchmark Generation Framework for Evaluating Frontier AI Models III: Implementing the Bacterial Biothreat Benchmark (B3) Dataset

✇cs.AI, q-bio.NC updates on arXiv.org
By: Gary Ackerman · Theodore Wilson · Zachary Kallenborn · Olivia Shoemaker · Anna Wetzel · Hayley Peterson · Abigail Danfora · Jenna LaTourette · Brandon Behlendorf · Douglas Clifford
10 December 2025 at 13:00
arXiv:2512.08459v1 Announce Type: cross Abstract: The potential for rapidly-evolving frontier artificial intelligence (AI) models, especially large language models (LLMs), to facilitate bioterrorism or access to biological weapons has generated significant policy, academic, and public concern. Both model developers and policymakers seek to quantify and mitigate any risk, with an important element of such efforts being the development of model benchmarks that can assess the biosecurity risk posed by a particular model. This paper discusses the pilot implementation of the Bacterial Biothreat Benchmark (B3) dataset. It is the third in a series of three papers describing an overall Biothreat Benchmark Generation (BBG) framework, with previous papers detailing the development of the B3 dataset. The pilot involved running the benchmarks through a sample frontier AI model, followed by human evaluation of model responses, and an applied risk analysis of the results along several dimensions. Overall, the pilot demonstrated that the B3 dataset offers a viable, nuanced method for rapidly assessing the biosecurity risk posed by a LLM, identifying the key sources of that risk and providing guidance for priority areas of mitigation priority.

Multi-domain performance analysis with scores tailored to user preferences

✇cs.AI, q-bio.NC updates on arXiv.org
By: S\'ebastien Pi\'erard · Adrien Deli\`ege · Marc Van Droogenbroeck
10 December 2025 at 13:00
arXiv:2512.08715v1 Announce Type: cross Abstract: The performance of algorithms, methods, and models tends to depend heavily on the distribution of cases on which they are applied, this distribution being specific to the applicative domain. After performing an evaluation in several domains, it is highly informative to compute a (weighted) mean performance and, as shown in this paper, to scrutinize what happens during this averaging. To achieve this goal, we adopt a probabilistic framework and consider a performance as a probability measure (e.g., a normalized confusion matrix for a classification task). It appears that the corresponding weighted mean is known to be the summarization, and that only some remarkable scores assign to the summarized performance a value equal to a weighted arithmetic mean of the values assigned to the domain-specific performances. These scores include the family of ranking scores, a continuum parameterized by user preferences, and that the weights to consider in the arithmetic mean depend on the user preferences. Based on this, we rigorously define four domains, named easiest, most difficult, preponderant, and bottleneck domains, as functions of user preferences. After establishing the theory in a general setting, regardless of the task, we develop new visual tools for two-class classification.

AI-powered virtual tissues from spatial proteomics for clinical diagnostics and biomedical discovery

✇cs.AI, q-bio.NC updates on arXiv.org
By: Johann Wenckstern · Eeshaan Jain · Yexiang Cheng · Benedikt von Querfurth · Kiril Vasilev · Matteo Pariset · Phil F. Cheng · Petros Liakopoulos · Olivier Michielin · Andreas Wicki · Gabriele Gut · Charlotte Bunne
10 December 2025 at 13:00
arXiv:2501.06039v2 Announce Type: replace-cross Abstract: Spatial proteomics technologies have transformed our understanding of complex tissue architecture in cancer but present unique challenges for computational analysis. Each study uses a different marker panel and protocol, and most methods are tailored to single cohorts, which limits knowledge transfer and robust biomarker discovery. Here we present Virtual Tissues (VirTues), a general-purpose foundation model for spatial proteomics that learns marker-aware, multi-scale representations of proteins, cells, niches and tissues directly from multiplex imaging data. From a single pretrained backbone, VirTues supports marker reconstruction, cell typing and niche annotation, spatial biomarker discovery, and patient stratification, including zero-shot annotation across heterogeneous panels and datasets. In triple-negative breast cancer, VirTues-derived biomarkers predict anti-PD-L1 chemo-immunotherapy response and stratify disease-free survival in an independent cohort, outperforming state-of-the-art biomarkers derived from the same datasets and current clinical stratification schemes.

OMNIGUARD: An Efficient Approach for AI Safety Moderation Across Languages and Modalities

✇cs.AI, q-bio.NC updates on arXiv.org
By: Sahil Verma · Keegan Hines · Jeff Bilmes · Charlotte Siska · Luke Zettlemoyer · Hila Gonen · Chandan Singh
10 December 2025 at 13:00
arXiv:2505.23856v2 Announce Type: replace-cross Abstract: The emerging capabilities of large language models (LLMs) have sparked concerns about their immediate potential for harmful misuse. The core approach to mitigate these concerns is the detection of harmful queries to the model. Current detection approaches are fallible, and are particularly susceptible to attacks that exploit mismatched generalization of model capabilities (e.g., prompts in low-resource languages or prompts provided in non-text modalities such as image and audio). To tackle this challenge, we propose Omniguard, an approach for detecting harmful prompts across languages and modalities. Our approach (i) identifies internal representations of an LLM/MLLM that are aligned across languages or modalities and then (ii) uses them to build a language-agnostic or modality-agnostic classifier for detecting harmful prompts. Omniguard improves harmful prompt classification accuracy by 11.57\% over the strongest baseline in a multilingual setting, by 20.44\% for image-based prompts, and sets a new SOTA for audio-based prompts. By repurposing embeddings computed during generation, Omniguard is also very efficient ($\approx\!120 \times$ faster than the next fastest baseline). Code and data are available at: https://github.com/vsahil/OmniGuard.

Development of a Hospital-at-Home Digital Twin for Patients With Frailty: Scoping Review

✇Journal of Medical Internet Research
By: Faiza Yahya · Matthew Cooper · Wahib Saif · Mohamad Kassem · Hamde Nazar
11 December 2025 at 05:15
Background: Increasing demand on healthcare systems requires innovative and transformative solutions to deliver efficient, high-quality care. One promising approach is Digital Twin (DT) technology, which leverages real time data to create dynamic virtual representations of a physical entity (individuals or space) to anticipate future scenarios and support care decisions. While DTs have been explored in various sectors, their application in Hospital at Home (HaH), which delivers acute level care in home environments, remains unexplored. Objective: This review bridges a critical knowledge gap and examines the existing evidence on DT-enabling tools for managing patients with frailty in home settings. This will identify the underpinning architectural components required to inform a HaH-DT system which can support clinical decision-making. Methods: Six electronic databases (Embase, MEDLINE, Cochrane CENTRAL, CINAHL, Web of Science and Scopus) were searched, along with grey literature, to identifying primary studies published in English, between January 2019 and September 2025. Included studies had to report on the monitoring or management of patients with frailty within their own home, and information was charted on a pre-defined data collection form to answer the research objectives. Review articles, protocols, and conference abstracts were excluded. Results: Sixty-nine reports were included, of which 54% (n=37) used quantitative approaches, and 36% (n=25) were pilot or feasibility studies. Reports were analysed for DT-enabling tools and systematically mapped across the proposed five-layered DT architecture: sensing, communication, storage, analytics, and visualisation. Taxonomies of DT layers, their interconnections, and the classifications of the types of data collected (e.g., about the patient, the home environment, the use of medical equipment) are presented. This evidence identifies DT-enabling tools used for a variety of functions and a range of sensing technologies that exist (e.g., passive sensing via wearables, active physiological sensors, ambient sensors to detect motion/environmental changes). The most prevalent modes of communication were wireless and network-based (n=36), with the majority using Bluetooth (n=12). This review highlights better understanding of data management, in particular secure storage, is required within local healthcare systems. The emerging potential of predictive and prescriptive analytics, which can enable clinicians to predict risk, support clinical decision-making, or activate alert-triggered health interventions were mapped. Existing evidence suggests analytics methods are currently largely descriptive with a lack of advanced methods such as prescriptive analytics to enable recommendations of an optimal course of action, and the absence of diagnostic analytics which can highlight why a situation has occurred. Reported DT-enabling tools demonstrate patient-centered benefits, including enhanced motivation, reassurance, and personalised care. However, concerns persist regarding device accuracy, user acceptability, and implications for carers and organisational workflows. Conclusions: This review is among the first to systematically map DT-enabling tools to inform a potential HaH-DT in patients with frailty and organised by a 5-layered conceptual model. Understanding these architectural layers provides the foundations to enable stakeholders advance research and development in areas where there are knowledge gaps and consider how a HaH DT can effectively operate within current healthcare systems. By leveraging technology-enabled care in complex home-based settings, there is great potential to deliver safer, personalised and timely care.
❌