❌

Reading view

↗️

XR-DT: Extended Reality-Enhanced Digital Twin for Agentic Mobile Robots

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Tianyi Wang Β· Jiseop Byeon Β· Ahmad Yehia Β· Huihai Wang Β· Yiming Xu Β· Tianyi Zeng Β· Ziran Wang Β· Junfeng Jiao Β· Christian Claudel
arXiv:2512.05270v1 Announce Type: cross Abstract: As mobile robots increasingly operate alongside humans in shared workspaces, ensuring safe, efficient, and interpretable Human-Robot Interaction (HRI) has become a pressing challenge. While substantial progress has been devoted to human behavior prediction, limited attention has been paid to how humans perceive, interpret, and trust robots' inferences, impeding deployment in safety-critical and socially embedded environments. This paper presents XR-DT, an eXtended Reality-enhanced Digital Twin framework for agentic mobile robots, that bridges physical and virtual spaces to enable bi-directional understanding between humans and robots. Our hierarchical XR-DT architecture integrates virtual-, augmented-, and mixed-reality layers, fusing real-time sensor data, simulated environments in the Unity game engine, and human feedback captured through wearable AR devices. Within this framework, we design an agentic mobile robot system with a unified diffusion policy for context-aware task adaptation. We further propose a chain-of-thought prompting mechanism that allows multimodal large language models to reason over human instructions and environmental context, while leveraging an AutoGen-based multi-agent coordination layer to enhance robustness and collaboration in dynamic tasks. Initial experimental results demonstrate accurate human and robot trajectory prediction, validating the XR-DT framework's effectiveness in HRI tasks. By embedding human intention, environmental dynamics, and robot cognition into the XR-DT framework, our system enables interpretable, trustworthy, and adaptive HRI.
  •  
  • ↗️
↗️

From Efficiency to Adaptivity: A Deeper Look at Adaptive Reasoning in Large Language Models

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Chao Wu Β· Baoheng Li Β· Mingchen Gao Β· Zhenyi Wang
arXiv:2511.10788v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have made reasoning a central benchmark for evaluating intelligence. While prior surveys focus on efficiency by examining how to shorten reasoning chains or reduce computation, this view overlooks a fundamental challenge: current LLMs apply uniform reasoning strategies regardless of task complexity, generating long traces for trivial problems while failing to extend reasoning for difficult tasks. This survey reframes reasoning through the lens of {adaptivity}: the capability to allocate reasoning effort based on input characteristics such as difficulty and uncertainty. We make three contributions. First, we formalize deductive, inductive, and abductive reasoning within the LLM context, connecting these classical cognitive paradigms with their algorithmic realizations. Second, we formalize adaptive reasoning as a control-augmented policy optimization problem balancing task performance with computational cost, distinguishing learned policies from inference-time control mechanisms. Third, we propose a systematic taxonomy organizing existing methods into training-based approaches that internalize adaptivity through reinforcement learning, supervised fine-tuning, and learned controllers, and training-free approaches that achieve adaptivity through prompt conditioning, feedback-driven halting, and modular composition. This framework clarifies how different mechanisms realize adaptive reasoning in practice and enables systematic comparison across diverse strategies. We conclude by identifying open challenges in self-evaluation, meta-reasoning, and human-aligned reasoning control.
  •  
  • ↗️
↗️

SciTrust 2.0: A Comprehensive Framework for Evaluating Trustworthiness of Large Language Models in Scientific Applications

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Emily Herron Β· Junqi Yin Β· Feiyi Wang
arXiv:2510.25908v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated transformative potential in scientific research, yet their deployment in high-stakes contexts raises significant trustworthiness concerns. Here, we introduce SciTrust 2.0, a comprehensive framework for evaluating LLM trustworthiness in scientific applications across four dimensions: truthfulness, adversarial robustness, scientific safety, and scientific ethics. Our framework incorporates novel, open-ended truthfulness benchmarks developed through a verified reflection-tuning pipeline and expert validation, alongside a novel ethics benchmark for scientific research contexts covering eight subcategories including dual-use research and bias. We evaluated seven prominent LLMs, including four science-specialized models and three general-purpose industry models, using multiple evaluation metrics including accuracy, semantic similarity measures, and LLM-based scoring. General-purpose industry models overall outperformed science-specialized models across each trustworthiness dimension, with GPT-o4-mini demonstrating superior performance in truthfulness assessments and adversarial robustness. Science-specialized models showed significant deficiencies in logical and ethical reasoning capabilities, along with concerning vulnerabilities in safety evaluations, particularly in high-risk domains such as biosecurity and chemical weapons. By open-sourcing our framework, we provide a foundation for developing more trustworthy AI systems and advancing research on model safety and ethics in scientific contexts.
  •  
  • ↗️
↗️

Multi-Agent Reinforcement Learning for Market Making: Competition without Collusion

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Ziyi Wang Β· Carmine Ventre Β· Maria Polukarov
arXiv:2510.25929v1 Announce Type: cross Abstract: Algorithmic collusion has emerged as a central question in AI: Will the interaction between different AI agents deployed in markets lead to collusion? More generally, understanding how emergent behavior, be it a cartel or market dominance from more advanced bots, affects the market overall is an important research question. We propose a hierarchical multi-agent reinforcement learning framework to study algorithmic collusion in market making. The framework includes a self-interested market maker (Agent~A), which is trained in an uncertain environment shaped by an adversary, and three bottom-layer competitors: the self-interested Agent~B1 (whose objective is to maximize its own PnL), the competitive Agent~B2 (whose objective is to minimize the PnL of its opponent), and the hybrid Agent~B$^\star$, which can modulate between the behavior of the other two. To analyze how these agents shape the behavior of each other and affect market outcomes, we propose interaction-level metrics that quantify behavioral asymmetry and system-level dynamics, while providing signals potentially indicative of emergent interaction patterns. Experimental results show that Agent~B2 secures dominant performance in a zero-sum setting against B1, aggressively capturing order flow while tightening average spreads, thus improving market execution efficiency. In contrast, Agent~B$^\star$ exhibits a self-interested inclination when co-existing with other profit-seeking agents, securing dominant market share through adaptive quoting, yet exerting a milder adverse impact on the rewards of Agents~A and B1 compared to B2. These findings suggest that adaptive incentive control supports more sustainable strategic co-existence in heterogeneous agent environments and offers a structured lens for evaluating behavioral design in algorithmic trading systems.
  •  
  • ↗️
↗️

Single-cell multi-omics analysis reveals cancer regulatory elements of transcriptional programs and clinical implications

βœ‡Omics In Lung
By: Xiaowei Tang Β· Qiaoling Zhang Β· Zichu Shen Β· Jian Xiao Β· Minghao Li Β· Xiangyan Meng Β· Chenyi Wang Β· Guangze Zhang Β· Anhang Liu Β· Yuxin Yin

Cell Death Dis. 2025 Oct 21;16(1):746. doi: 10.1038/s41419-025-08060-7.

ABSTRACT

The regulatory mechanisms governing transcriptional programs in the cancer genome remain elusive, particularly those concerning cell-type specificity. We carefully curated single-cell assay for transposase-accessible chromatin sequencing (scATAC-seq) and single-cell RNA sequencing (scRNA-seq) data from eight distinct carcinoma tissues, including breast, skin, colon, endometrium, lung, ovary, liver, and kidney. Using single-cell multi-omics analysis, we identified extensive open chromatin regions and constructed peak-gene link networks, which can reveal distinct cancer gene regulation and genetic risks. We further explored conserved epigenetic regulation across cell types within cancer and elucidated their functional implications. Moreover, we identified cell-type-associated transcription factors (TFs) that regulate key cellular functions, such as the TEAD family of TFs, which widely control cancer-related signaling pathways in tumor cells. In colon cancer, we further identified tumor-specific TFs that are more highly activated in tumor cells than in normal epithelial cells, including CEBPG, LEF1, SOX4, TCF7, and TEAD4, which are pivotal in driving malignant transcriptional programs and represent potential therapeutic targets, as corroborated by single-cell sequencing data from multiple sources and in vitro experiments. Our findings provide a comprehensive understanding of the regulatory dynamics underlying carcinomas and offer valuable insights into potential therapeutic interventions.

PMID:41120274 | PMC:PMC12541060 | DOI:10.1038/s41419-025-08060-7

  •  
  • ↗️
↗️

Using Large Language Models to Assess the Consistency of Randomized Controlled Trials on AI Interventions With CONSORT-AI: Cross-Sectional Survey

βœ‡Journal of Medical Internet Research
By: Xufei Luo Β· Zeming Li Β· Zhenhua Yang Β· Bingyi Wang Β· Yanfang Ma Β· Fengxian Chen Β· Qi Wang Β· Long Ge Β· James Zou Β· Lu Zhang Β· Yaolong Chen Β· Zhaoxiang Bian
Background: Chatbots based on large language models (LLMs) have shown promise in evaluating the consistency of research. Previously, researchers used LLM to assess if randomized controlled trial (RCT) abstracts adhered to the CONSORT-Abstract guidelines. However, the consistency of artificial intelligence (AI) interventional RCTs align with the CONSORT-AI standards by LLMs remains unclear. Objective: The aim of this study is to identify the consistency of randomized controlled trials on AI interventions with CONSORT-AI using chatbots based on LLMs. Methods: This cross-sectional study employed six LLM models to assess the consistency of RCTs on AI interventions. The sample selection is based on articles published in JAMA Network Open, which included a total of 41 RCTs. All queries were submitted to LLMs through an API interface with a temperature setting of 0 to ensure deterministic responses. One researcher posed the questions to each model, while another independently verified the responses for validity before recording the results. The Overall Consistency Score (OCS), recall, inter-rater reliability and consistency of contents were analyzed. Results: We found gpt-4-0125-preview has the best average OCS (86.5%, 95%CI: 82.5%-90.5% and 81.6%, 95% CI: 77.6%-85.6%), followed by gpt-4-1106-preview(80.3%, 95%CI: 76.3%-84.3% and 78.0%, 95% CI: 74.0%-82.0%). The model with the worst average OCS is gpt-3.5-turbo-0125 (61.9%, 95%CI: 57.9%-65.9% and 63.0%, 95% CI: 59.0%-67.0%). Among the 11 unique items of CONSORT-AI, Item 2 (β€œState the inclusion and exclusion criteria at the level of the input data”) received the poorest overall evaluation across six models, with an average OCS of 48.8%. For other items, those with an average OCS greater than 80% across the six models included Items 1, 5, 8, and 9. Conclusions: GPT-4 variants demonstrate strong performance in assessing the consistency of RCTs with CONSORT-AI. Nonetheless, refining the prompts could enhance the precision and consistency of the outcomes. While AI tools like GPT-4 variants are valuable, they are not yet fully autonomous in addressing complex and nuanced tasks such as adherence to CONSORT-AI standards. Therefore, integrating AI with higher levels of human supervision and expertise will be crucial to ensuring more reliable and efficient evaluations, ultimately advancing the quality of medical research.
  •  
  • ↗️
↗️

Decoding per- and polyfluoroalkyl substances (PFAS) in hepatocellular carcinoma: a multi-omics and computational toxicology approach

βœ‡Omics in Hepatocellular
By: Yanggang Hong Β· Deqi Wang Β· Zeyu Liu Β· Yuxin Chen Β· Yi Wang Β· Jiajun Li

J Transl Med. 2025 May 2;23(1):504. doi: 10.1186/s12967-025-06517-z.

ABSTRACT

BACKGROUND: Per- and polyfluoroalkyl substances (PFAS), particularly perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS), are synthetic chemicals known for their widespread use and environmental persistence. These compounds have been increasingly linked to hepatotoxicity and the development of hepatocellular carcinoma (HCC). However, the molecular mechanisms by which PFAS contribute to HCC remain underexplored.

METHODS: This study employs a multi-omics approach that combines network toxicology, integrated machine learning, single-cell RNA sequencing, spatial transcriptomics, experimental validation, and molecular docking simulations to uncover the mechanisms through which PFAS exposure drives HCC. We analyzed publicly available transcriptomic data from several HCC cohorts and used differential gene expression analysis to identify targets associated with both PFAS exposure and HCC. We constructed a protein-protein interaction (PPI) network and a survival risk model, the PFAS-related HCC signature (PFASRHSig), based on integrated machine learning to identify prognostic biomarkers, with the goal of identifying core targets of PFAS in HCC progression and prognosis. RT-qPCR and immunohistochemical (IHC) staining were used to validate the expression levels of the targets in both tumor and normal tissues. Molecular docking simulations were conducted to assess the binding affinities between PFAS compounds and selected target proteins.

RESULTS: Functional enrichment studies revealed that PFAS targets were associated with metabolic signaling pathways, which are actively involved in lipid, glucose, drug metabolism, etc. Through integrated machine learning and PPI network analysis, we identified six genes, APOA1, ESR1, IGF1, PPARGC1A, SERPINE1, and PON1, that serve as core targets of PFAS in both HCC progression and prognosis. These targets were further validated via bulk RNA-seq, single-cell RNA-seq, and spatial transcriptomics, which revealed differential expression patterns across various cell types in the HCC tumor microenvironment. The results of RT-qPCR and IHC staining were consistent with the in silico findings. Molecular docking simulations revealed strong binding affinities between PFAS compounds and these core targets, supporting their potential roles in PFAS-induced hepatocarcinogenesis.

CONCLUSIONS: Our study highlights key molecular targets and pathways involved in PFAS-induced liver carcinogenesis and proposes a robust survival risk model (PFASRHSig) for HCC. These findings provide new insights into PFAS toxicity mechanisms and offer potential therapeutic targets for mitigating the health risks associated with PFAS exposure. Collectively, our findings help in advancing clinical applications by providing insights into disease mechanisms and potential therapeutic interventions.

PMID:40317014 | PMC:PMC12049027 | DOI:10.1186/s12967-025-06517-z

  •  
  • ↗️
↗️

A deep learning framework for <em>in silico</em> screening of anticancer drugs at the single-cell level

βœ‡(Multiomics OR Omics) AND (Pancreatic)
By: Peijing Zhang Β· Xueyi Wang Β· Xufeng Cen Β· Qi Zhang Β· Yuting Fu Β· Yuqing Mei Β· Xinru Wang Β· Renying Wang Β· Jingjing Wang Β· Hongwei Ouyang Β· Tingbo Liang Β· Hongguang Xia Β· Xiaoping Han Β· Guoji Guo

Natl Sci Rev. 2024 Dec 10;12(2):nwae451. doi: 10.1093/nsr/nwae451. eCollection 2025 Feb.

ABSTRACT

Tumor heterogeneity plays a pivotal role in tumor progression and resistance to clinical treatment. Single-cell RNA sequencing (scRNA-seq) enables us to explore heterogeneity within a cell population and identify rare cell types, thereby improving our design of targeted therapeutic strategies. Here, we use a pan-cancer and pan-tissue single-cell transcriptional landscape to reveal heterogeneous expression patterns within malignant cells, precancerous cells, as well as cancer-associated stromal and endothelial cells. We introduce a deep learning framework named Shennong for in silico screening of anticancer drugs for targeting each of the landscape cell clusters. Utilizing Shennong, we could predict individual cell responses to pharmacologic compounds, evaluate drug candidates' tissue damaging effects, and investigate their corresponding action mechanisms. Prioritized compounds in Shennong's prediction results include FDA-approved drugs currently undergoing clinical trials for new indications, as well as drug candidates reporting anti-tumor activity. Furthermore, the tissue damaging effect prediction aligns with documented injuries and terminated discovery events. This robust and explainable framework has the potential to accelerate the drug discovery process and enhance the accuracy and efficiency of drug screening.

PMID:39872221 | PMC:PMC11771446 | DOI:10.1093/nsr/nwae451

  •  
  • ↗️
There are no more articles
❌