id
large_stringlengths 9
16
| title
large_stringlengths 1
382
| abstract
large_stringlengths 3
6.09k
| publish_date
date32 | update_date
date32 | categories
large listlengths 1
13
| authors
large_stringlengths 3
62.8k
|
|---|---|---|---|---|---|---|
2512.22392
|
iOSPointMapper: RealTime Pedestrian and Accessibility Mapping with Mobile AI
|
Accurate, up-to-date sidewalk data is essential for building accessible and inclusive pedestrian infrastructure, yet current approaches to data collection are often costly, fragmented, and difficult to scale. We introduce iOSPointMapper, a mobile application that enables real-time, privacy-conscious sidewalk mapping on the ground, using recent-generation iPhones and iPads. The system leverages on-device semantic segmentation, LiDAR-based depth estimation, and fused GPS/IMU data to detect and localize sidewalk-relevant features such as traffic signs, traffic lights and poles. To ensure transparency and improve data quality, iOSPointMapper incorporates a user-guided annotation interface for validating system outputs before submission. Collected data is anonymized and transmitted to the Transportation Data Exchange Initiative (TDEI), where it integrates seamlessly with broader multimodal transportation datasets. Detailed evaluations of the system's feature detection and spatial mapping performance reveal the application's potential for enhanced pedestrian mapping. Together, these capabilities offer a scalable and user-centered approach to closing critical data gaps in pedestrian
| 2025-12-26
| 2025-12-30
|
[
"cs.CV"
] |
Himanshu Naidu, Yuxiang Zhang, Sachin Mehta, Anat Caspi
|
2512.21983
|
Bab_Sak Robotic Intubation System (BRIS): A Learning-Enabled Control Framework for Safe Fiberoptic Endotracheal Intubation
|
Endotracheal intubation is a critical yet technically demanding procedure, with failure or improper tube placement leading to severe complications. Existing robotic and teleoperated intubation systems primarily focus on airway navigation and do not provide integrated control of endotracheal tube advancement or objective verification of tube depth relative to the carina. This paper presents the Robotic Intubation System (BRIS), a compact, human-in-the-loop platform designed to assist fiberoptic-guided intubation while enabling real-time, objective depth awareness. BRIS integrates a four-way steerable fiberoptic bronchoscope, an independent endotracheal tube advancement mechanism, and a camera-augmented mouthpiece compatible with standard clinical workflows. A learning-enabled closed-loop control framework leverages real-time shape sensing to map joystick inputs to distal bronchoscope tip motion in Cartesian space, providing stable and intuitive teleoperation under tendon nonlinearities and airway contact. Monocular endoscopic depth estimation is used to classify airway regions and provide interpretable, anatomy-aware guidance for safe tube positioning relative to the carina. The system is validated on high-fidelity airway mannequins under standard and difficult airway configurations, demonstrating reliable navigation and controlled tube placement. These results highlight BRIS as a step toward safer, more consistent, and clinically compatible robotic airway management.
| 2025-12-26
| 2025-12-29
|
[
"cs.RO"
] |
Saksham Gupta, Sarthak Mishra, Arshad Ayub, Kamran Farooque, Spandan Roy, Babita Gupta
|
2512.13478
|
Non-Resolution Reasoning (NRR): A Computational Framework for Contextual Identity and Ambiguity Preservation
|
Current AI systems exhibit a fundamental limitation: they resolve ambiguity prematurely. This premature semantic collapse--collapsing multiple valid interpretations into single outputs--stems from classical identity assumptions in neural architectures. We propose Non-Resolution Reasoning (NRR), treating ambiguity retention as a valid reasoning mode. NRR introduces three principles: (1) Non-Identity ($A \neq A$)--the same symbol refers to different entities across contexts; (2) Approximate Identity ($A \approx A$)--entities share partial overlap without being identical; (3) Non-Resolution--conflicting interpretations coexist without forced convergence. We formalize these through Multi-Vector Embeddings, Non-Collapsing Attention, and Contextual Identity Tracking (CIT). Functional verification via Turn 1 Entropy measurement shows NRR-lite maintains high entropy ($H = 0.63$) at ambiguous turns while standard architectures collapse early ($H = 0.10$), demonstrating that NRR preserves interpretive flexibility until context arrives. The question is not whether AI should resolve ambiguity, but when, how, and under whose control.
| 2025-12-26
| 2025-12-29
|
[
"cs.CL",
"cs.AI",
"cs.LG"
] |
Kei Saito
|
2511.20663
|
MTTR-A: Measuring Cognitive Recovery Latency in Multi-Agent Systems
|
Reliability in multi-agent systems (MAS) built on large language models is increasingly limited by cognitive failures rather than infrastructure faults. Existing observability tools describe failures but do not quantify how quickly distributed reasoning recovers once coherence is lost. We introduce MTTR-A (Mean Time-to-Recovery for Agentic Systems), a runtime reliability metric that measures cognitive recovery latency in MAS. MTTR-A adapts classical dependability theory to agentic orchestration, capturing the time required to detect reasoning drift and restore coherent operation. We further define complementary metrics, including MTBF and a normalized recovery ratio (NRR), and establish theoretical bounds linking recovery latency to long-run cognitive uptime. Using a LangGraph-based benchmark with simulated drift and reflex recovery, we empirically demonstrate measurable recovery behavior across multiple reflex strategies. This work establishes a quantitative foundation for runtime cognitive dependability in distributed agentic systems.
| 2025-12-26
| 2025-12-30
|
[
"cs.MA",
"cs.AI",
"cs.SY",
"eess.SY"
] |
Barak Or
|
2212.04195
|
A Paradigm Shift in Human Neuroscience Research: Progress, Prospects, and a Proof of Concept for Population Neuroscience
|
Recent advances and reflections on reproducible human neuroscience, especially brain-wide association studies (BWAS) leveraging large datasets, have led to divergent and sometimes opposing views on research practices and priorities. The debates span multiple dimensions. Shifts along these axes have fractured consensus and further fragmented an already heterogeneous field of cognitive neuroscience. Here, we sketch a holistic and integrative response grounded in population neuroscience, organized around a closed-loop "design-analysis-interpretation" research cycle that aims to build consensus while bridging these divides. Our central claim is that population neuroscience offers a unique population-level vantage point for identifying general principles, characterizing inter-individual variabilities, and benchmarking intra-individual changes, thereby providing a supportive framework for small-scale, mechanism-focused studies at the individual level and allowing them to co-evolve with population-level studies. Population neuroscience is not simply about providing larger N for BWAS; its deeper goal is to accumulate a family of cross-scale priors and shared infrastructures that can support design, analysis, and interpretation of human neuroscience for decades to come. In this sense, we outline a "third-generation" view of population neuroscience that reorients the field from amassing isolated associations toward building integrative reference frameworks for future mechanistic and translational work.
| 2025-12-26
| 2025-12-30
|
[
"q-bio.NC",
"q-bio.QM",
"stat.ME"
] |
Zi-Xuan Zhou, Xi-Nian Zuo
|
2512.22001
|
Variational Quantum Eigensolver for Real-World Finance: Scalable Solutions for Dynamic Portfolio Optimization Problems
|
We present a scalable, hardware-aware methodology for extending the Variational Quantum Eigensolver (VQE) to large, realistic Dynamic Portfolio Optimization (DPO) problems. Building on the scaling strategy from our previous work, where we tailored a VQE workflow to both the DPO formulation and the target QPU, we now put forward two significant advances. The first is the implementation of the Ising Sample-based Quantum Configuration Recovery (ISQR) routine, which improves solution quality in Quadratic Unconstrained Binary Optimization problems. The second is the use of the VQE Constrained method to decompose the optimization task, enabling us to handle DPO instances with more variables than the available qubits on current hardware. These advances, which are broadly applicable to other optimization problems, allow us to address a portfolio with a size relevant to the financial industry, consisting of up to 38 assets and covering the full Spanish stock index (IBEX 35). Our results, obtained on a real Quantum Processing Unit (IBM Fez), show that this tailored workflow achieves financial performance on par with classical methods while delivering a broader set of high-quality investment strategies, demonstrating a viable path towards obtaining practical advantage from quantum optimization in real financial applications.
| 2025-12-26
| 2025-12-29
|
[
"quant-ph",
"q-fin.CP",
"q-fin.PM"
] |
Irene De León, Danel Arias, Manuel MartÃn-Cordero, MarÃa Esperanza Molina, Pablo Serrano, Senaida Hernández-Santana, Miguel Ãngel Jiménez Herrera, Joana Fraxanet, Ginés Carrascal, Escolástico Sánchez, Inmaculada Posadillo, Ãlvaro Nodar
|
2308.15451
|
Metawisdom of the Crowd: Experimental Evidence of Crowd Accuracy Through Diverse Choices of Decision Aids
|
The provision of information can improve individual judgments but also fail to make group decisions more accurate; if individuals choose to attend to the same information in the same manner, the predictive diversity that enables crowd wisdom may be lost. Decision support systems, from search engines to business intelligence platforms, present individuals with decision aids -- relevant information, interpretative frames, or heuristics -- to enhance the quality and speed of decision-making but potentially influence judgments through the selective presentation of information and interpretative frames. We describe decision-making as often containing two decisions: the choice of decision aids followed by the primary decision, and define \textit{metawisdom of the crowd} as any pattern by which individuals' choice of aids leads to higher crowd accuracy than equal assignment to the same aids, a comparison that accounts for the information content of the aids. The theoretical model accounting for aid bias and variance shows that an optimal distribution of aid usage can produce metawisdom based on the characteristics of aids within a collection. Three studies -- two estimation tasks (N=900, 728) and the nowcasting of inflation (N=1,956; across three collections) -- support this claim. Metawisdom emerges from the use of diverse aids, not through widespread use of the aids that induce the most accurate estimates. Thus, the microfoundations of crowd wisdom appear in the first choice, suggesting crowd wisdom can be robust in information choice problems. Given the implications for collective decision-making, the insights warrant future research investigations into the nature and use of decision aids.
| 2025-12-26
| 2025-12-29
|
[
"econ.GN",
"q-fin.EC"
] |
Jon Atwell, Marlon Twyman
|
2512.22065
|
StreamAvatar: Streaming Diffusion Models for Real-Time Interactive Human Avatars
|
Real-time, streaming interactive avatars represent a critical yet challenging goal in digital human research. Although diffusion-based human avatar generation methods achieve remarkable success, their non-causal architecture and high computational costs make them unsuitable for streaming. Moreover, existing interactive approaches are typically limited to head-and-shoulder region, limiting their ability to produce gestures and body motions. To address these challenges, we propose a two-stage autoregressive adaptation and acceleration framework that applies autoregressive distillation and adversarial refinement to adapt a high-fidelity human video diffusion model for real-time, interactive streaming. To ensure long-term stability and consistency, we introduce three key components: a Reference Sink, a Reference-Anchored Positional Re-encoding (RAPR) strategy, and a Consistency-Aware Discriminator. Building on this framework, we develop a one-shot, interactive, human avatar model capable of generating both natural talking and listening behaviors with coherent gestures. Extensive experiments demonstrate that our method achieves state-of-the-art performance, surpassing existing approaches in generation quality, real-time efficiency, and interaction naturalness. Project page: https://streamavatar.github.io .
| 2025-12-26
| 2025-12-29
|
[
"cs.CV",
"cs.AI",
"cs.HC"
] |
Zhiyao Sun, Ziqiao Peng, Yifeng Ma, Yi Chen, Zhengguang Zhou, Zixiang Zhou, Guozhen Zhang, Youliang Zhang, Yuan Zhou, Qinglin Lu, Yong-Jin Liu
|
2512.21857
|
Fast Inference of Visual Autoregressive Model with Adjacency-Adaptive Dynamical Draft Trees
|
Autoregressive (AR) image models achieve diffusion-level quality but suffer from sequential inference, requiring approximately 2,000 steps for a 576x576 image. Speculative decoding with draft trees accelerates LLMs yet underperforms on visual AR models due to spatially varying token prediction difficulty. We identify a key obstacle in applying speculative decoding to visual AR models: inconsistent acceptance rates across draft trees due to varying prediction difficulties in different image regions. We propose Adjacency-Adaptive Dynamical Draft Trees (ADT-Tree), an adjacency-adaptive dynamic draft tree that dynamically adjusts draft tree depth and width by leveraging adjacent token states and prior acceptance rates. ADT-Tree initializes via horizontal adjacency, then refines depth/width via bisectional adaptation, yielding deeper trees in simple regions and wider trees in complex ones. The empirical evaluations on MS-COCO 2017 and PartiPrompts demonstrate that ADT-Tree achieves speedups of 3.13xand 3.05x, respectively. Moreover, it integrates seamlessly with relaxed sampling methods such as LANTERN, enabling further acceleration. Code is available at https://github.com/Haodong-Lei-Ray/ADT-Tree.
| 2025-12-26
| 2025-12-29
|
[
"cs.CV"
] |
Haodong Lei, Hongsong Wang, Xin Geng, Liang Wang, Pan Zhou
|
2512.22399
|
Space AI: Leveraging Artificial Intelligence for Space to Improve Life on Earth
|
Artificial Intelligence (AI) is transforming domains from healthcare and agriculture to finance and industry. As progress on Earth meets growing constraints, the next frontier is outer space, where AI can enable autonomous, resilient operations under extreme uncertainty and limited human oversight. This paper introduces Space AI as a unified interdisciplinary field at the intersection of artificial intelligence and space science and technology. We consolidate historical developments and contemporary progress, and propose a systematic framework that organises Space AI into four mission contexts: 1 AI on Earth, covering intelligent mission planning, spacecraft design optimisation, simulation, and ground-based data analytics; 2 AI in Orbit, focusing on satellite and station autonomy, space robotics, on-board/near-real-time data processing, communication optimisation, and orbital safety; (3) AI in Deep Space, enabling autonomous navigation, adaptive scientific discovery, resource mapping, and long-duration human-AI collaboration under communication constraints; and 4 AI for Multi-Planetary Life, supporting in-situ resource utilisation, habitat and infrastructure construction, life-support and ecological management, and resilient interplanetary networks. Ultimately, Space AI can accelerate humanity's capability to explore and operate in space, while translating advances in sensing, robotics, optimisation, and trustworthy AI into broad societal impact on Earth.
| 2025-12-26
| 2025-12-30
|
[
"astro-ph.IM",
"cs.AI",
"physics.space-ph"
] |
Ziyang Wang
|
2512.15144
|
MCPZoo: A Large-Scale Dataset of Runnable Model Context Protocol Servers for AI Agent
|
Model Context Protocol (MCP) enables agents to interact with external tools, yet empirical research on MCP is hindered by the lack of large-scale, accessible datasets. We present MCPZoo, the largest and most comprehensive dataset of MCP servers collected from multiple public sources, comprising 129,059 servers (56,053 distinct). MCPZoo includes 16,356 server instances that have been deployed and verified as runnable and interactable, supporting realistic experimentation beyond static analysis. The dataset provides unified metadata and access interfaces, enabling systematic exploration and interaction without manual deployment effort. MCPZoo is released as an open and accessible resource to support research on MCP-based systems and security analysis.
| 2025-12-26
| 2025-12-29
|
[
"cs.CR"
] |
Mengying Wu, Pei Chen, Geng Hong, Baichao An, Jinsong Chen, Binwang Wan, Xudong Pan, Jiarun Dai, Min Yang
|
2512.21967
|
BLEST: Blazingly Efficient BFS using Tensor Cores
|
Breadth-First Search (BFS) is a fundamental graph kernel that underpins a wide range of applications. While modern GPUs provide specialised Matrix-Multiply-Accumulate (MMA) units, e.g., Tensor Cores (TC), with extremely high throughput, they target dense operations, making it non-trivial to exploit them for irregular, unstructured graph computations. In particular, fully utilising them for a BFS requires an efficient mapping of the edge operations onto TCs while avoiding redundancy, load imbalance, and synchronisation. We present BLEST, a TC-accelerated framework that reformulates the pull-based BFS pipeline around a bitmap-oriented structure and a carefully engineered execution layout. BLEST introduces Binarised Virtual Slice Sets (BVSS) to enforce warp-level load balancing and to eliminate frontier-oblivious work assignment. To improve both memory efficiency and update locality across diverse graphs, we apply two complementary graph reordering strategies: a compression-oriented ordering for social-like graphs and a bandwidth-reducing ordering for non-social graphs. At the compute level, we develop a batched SpMSpV multiplication pattern that uses the bitwise TC tiles to handle dot products without wasting output entries, thereby reducing the number of required MMA calls. Finally, BLEST combines kernel fusion with a lazy vertex update scheme to reduce host-side synchronisation, mitigate atomic overheads, and improve cache locality. Experiments show that BLEST delivers, on average, $3.58\times$, $4.64\times$ and $4.9\times$ speedup over BerryBees, Gunrock, and GSWITCH, respectively, across a broad set of real-world graphs.
| 2025-12-26
| 2025-12-29
|
[
"cs.DC",
"cs.DS"
] |
Deniz Elbek, Kamer Kaya
|
2512.21813
|
Organizational Learning in Industry 4.0: Applying Crossan's 4I Framework with Double Loop Learning
|
The Advanced Dynamic Security Learning (DSL) Process Model is an Industry 4.0 cybersecurity incident response architecture proposed in this paper. This model addresses proactive and reflective cybersecurity governance across complex cyber-physical systems by combining Argyris and Schön's double-loop learning theory with Crossan's 4I organizational learning framework. Given that 65% of industrial companies suffer ransomware attacks annually and many of them lack cybersecurity awareness, this reveals the gravity of cyber threats. Feedforward and feedback learning loops in this paradigm help promote strategic transformation and ongoing growth. The DSL model helps Industry 4.0 organizations adapt to growing challenges posed by the projected 18.8 billion IoT devices by bridging operational obstacles and promoting systemic resilience. This research presents a scalable, methodical cybersecurity maturity approach based on a comprehensive analysis of the literature and a qualitative study.
| 2025-12-26
| 2025-12-29
|
[
"cs.CR"
] |
Nimra Akram, Atif Ahmad, Sean B Maynard
|
2512.21902
|
Explainable Statute Prediction via Attention-based Model and LLM Prompting
|
In this paper, we explore the problem of automatic statute prediction where for a given case description, a subset of relevant statutes are to be predicted. Here, the term "statute" refers to a section, a sub-section, or an article of any specific Act. Addressing this problem would be useful in several applications such as AI-assistant for lawyers and legal question answering system. For better user acceptance of such Legal AI systems, we believe the predictions should also be accompanied by human understandable explanations. We propose two techniques for addressing this problem of statute prediction with explanations -- (i) AoS (Attention-over-Sentences) which uses attention over sentences in a case description to predict statutes relevant for it and (ii) LLMPrompt which prompts an LLM to predict as well as explain relevance of a certain statute. AoS uses smaller language models, specifically sentence transformers and is trained in a supervised manner whereas LLMPrompt uses larger language models in a zero-shot manner and explores both standard as well as Chain-of-Thought (CoT) prompting techniques. Both these models produce explanations for their predictions in human understandable forms. We compare statute prediction performance of both the proposed techniques with each other as well as with a set of competent baselines, across two popular datasets. Also, we evaluate the quality of the generated explanations through an automated counter-factual manner as well as through human evaluation.
| 2025-12-26
| 2025-12-29
|
[
"cs.CL"
] |
Sachin Pawar, Girish Keshav Palshikar, Anindita Sinha Banerjee, Nitin Ramrakhiyani, Basit Ali
|
2512.21907
|
SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?
|
Spatial transcriptomics assays are rapidly increasing in scale and complexity, making computational analysis a major bottleneck in biological discovery. Although frontier AI agents have improved dramatically at software engineering and general data analysis, it remains unclear whether they can extract biological insight from messy, real-world spatial datasets. We introduce SpatialBench, a benchmark of 146 verifiable problems derived from practical spatial analysis workflows spanning five spatial technologies and seven task categories. Each problem provides a snapshot of experimental data immediately prior to an analysis step and a deterministic grader that evaluates recovery of a key biological result. Benchmark data on frontier models shows that base model accuracy remains low (20-38% across model families), with strong model-task and model-platform interactions. Harness design has a large empirical effect on performance, indicating that tools, prompts, control flow, and execution environment should be evaluated and improved as first-class objects. SpatialBench serves both as a measurement tool and a diagnostic lens for developing agents that can interact with real spatial datasets faithfully, transparently, and reproducibly.
| 2025-12-26
| 2025-12-29
|
[
"cs.AI"
] |
Kenny Workman, Zhen Yang, Harihara Muralidharan, Hannah Le
|
2512.21844
|
A Cohomological Framework for Topological Phases from Momentum-Space Crystallographic Groups
|
Crystallographic groups are conventionally studied in real space to characterize crystal symmetries. Recent work has recognized that when these symmetries are realized projectively, momentum space inherently accommodates nonsymmorphic symmetries, thereby evoking the concept of \textit{momentum-space crystallographic groups} (MCGs). Here, we reveal that the cohomology of MCGs encodes fundamental data of crystalline topological band structures. Specifically, the collection of second cohomology groups, $H^2(Î_F,\mathbb{Z})$, for all MCGs $Î_F$, provides an exhaustive classification of Abelian crystalline topological insulators, serving as an effective approximation to the full crystalline topological classification. Meanwhile, the third cohomology groups $H^3(Î_F,\mathbb{Z})$ across all MCGs exhaustively classify all possible twistings of point-group actions on the Brillouin torus, essential data for twisted equivariant K-theory. Furthermore, we establish the isomorphism $H^{n+1}(Î_F,\mathbb{Z})\cong H^n\big(Î_F,\operatorname{\mathcal{F}}(\mathbb{R}^d_F,U(1))\big)$ for $ n\ge 1$, where $\operatorname{\mathcal{F}}(\mathbb{R}^d_F,U(1))$ denotes the space of continuous $U(1)$-valued functions on the $d$D momentum space $\mathbb{R}^d_F$. The case $n=1$ yields a complete set of topological invariants formulated in purely algebraic terms, which differs fundamentally from the conventional formulation in terms of differential forms. The case $n=2$, analogously, provides a fully algebraic description for all such twistings. Thus, the cohomological theory of MCGs serves as a key technical framework for analyzing crystalline topological phases within the general setting of projective symmetry.
| 2025-12-26
| 2025-12-29
|
[
"cond-mat.mes-hall",
"cond-mat.str-el",
"math-ph",
"math.MP",
"quant-ph"
] |
T. R. Liu, Zheng Zhang, Y. X. Zhao
|
2512.21976
|
Finite Groups of Random Walks in the Quarter Plane and Periodic $4$-bar Links
|
We solve two long standing open problems, one from probability theory formulated by Malyshev in 1970 and another one from a crossroad of geometry and dynamics, going back to Darboux in 1879. The Malyshev problem is of finding effective, explicit necessary and sufficient conditions in the closed form to characterize all random walks in the quarter plane with a finite group of the random walk of order $2n$, for all $n\ge 2$, in the generic case where the underlining biquadratic is an elliptic curve. Until now, the results were known only for $n=2, 3, 4$ and were obtained using ad-hoc methods developed separately for each of the three cases. We provide a method that solves the problem for all $n$ and in a unified way. We also consider situations with singular biquadratics. Further, we establish a new two-way relationship between \emph{diagonal} random walks in the quarter plane and $4$-bar links. We describe all $n$-periodic Darboux transformations for $4$-bar link problems for all $n\ge 2$, thus completely solving the Darboux problem, that he solved for $n=2$. We introduce \emph{$k$ semi-periodicity} as a novel and natural type of periodicity of the Darboux transformations, where after $k$ iterations of the Darboux transformation, a polygonal configuration maps to a congruent one, but of opposite orientation. By introducing new objects, \emph{the secondary $(2-2)$-correspondence} and the related \emph{secondary cubic} of the centrally-symmetric biquadratics, we provide necessary and sufficient conditions for $k$-semi-periodicity for $4$-bar links for all $k\ge 2$ in an explicit closed form.
| 2025-12-26
| 2025-12-29
|
[
"math.AG",
"math.CO",
"math.DS",
"math.MG",
"math.PR"
] |
Vladimir DragoviÄ, Milena RadnoviÄ
|
2512.22320
|
A Time-Symmetric Variational Formulation of Quantum Mechanics with Emergent Schrödinger Dynamics and Objective Boundary Randomness
|
We present a time-symmetric variational formulation of nonrelativistic quantum mechanics in which Schrödinger dynamics and a Bohm-type guidance law arise as emergent Euler-Lagrange optimality conditions rather than postulates. The formulation is expressed in terms of probability density and current fields subject to a continuity constraint and two-time boundary conditions. A Fisher-information regularization term generates the quantum potential, yielding the Schrödinger equation when the optimality system is expressed in complex form. Unlike standard Bohmian mechanics, which requires an auxiliary Quantum Equilibrium Hypothesis ($P = |Ï|^2$), our primal-dual formulation satisfies the Born rule by construction. The trajectories emerge not from an external guidance wave, but as the unique hydrodynamic flow minimizing the Fisher-regularized action between two-time boundary constraints. Deterministic trajectories thus emerge only as effective, coarse-grained descriptions, with randomness entering objectively at the interface of boundary constraints.
| 2025-12-26
| 2025-12-30
|
[
"quant-ph"
] |
Lance H. Carter
|
2512.21869
|
Theoretical perspectives on charge dynamics in high-temperature cuprate superconductors
|
We review recent theoretical progress on the charge dynamics of doped carriers in high-temperature cuprate superconductors. Advances in this field have clarified that doped charges in cuprates exhibit remarkably rich collective behavior, governed by the combined effects of strong electronic correlations, the intrinsic layered crystal structure, and long-range Coulomb interaction. First, the emergence of acousticlike plasmons has been firmly established through quantitative analyses of resonant inelastic x-ray scattering (RIXS) spectra based on the t-J-V model -- an extension of the conventional t-J model that incorporates the layered crystal structure and the long-range Coulomb interaction V. These acousticlike plasmons arise near the in-plane momentum q=(0,0) and possess characteristic energies far below the well-known ~ 1 eV optical plasmon. This behavior is found to be universal across both hole- and electron-doped cuprates, including multilayer systems. Second, in electron-doped cuprates, a pronounced tendency toward d-wave bond-charge order develops near q=(0.5pi, 0), as revealed by resonant x-ray scattering (RXS) and RIXS. As a result, the charge dynamics acquires a dual structure, in which low-energy bond-charge excitations coexist with relatively high-energy plasmons. Third, analogous signatures of charge-order tendency have also been reported in hole-doped cuprates. However, a direct application of the d-wave bond-charge-order framework fails to account for experimental observations. Similarly, the charge-stripe order in La-based cuprates remains unresolved within existing theoretical approaches. Assuming that mobile carriers behave in a largely universal manner across electron- and hole-doped systems, we discuss a possible scenario that may reconcile these diverse experimental findings.
| 2025-12-26
| 2025-12-29
|
[
"cond-mat.str-el",
"cond-mat.supr-con"
] |
Hiroyuki Yamase
|
2509.05607
|
CC-GSEO-Bench: A Content-Centric Benchmark for Measuring Source Influence in Generative Search Engines
|
Generative Search Engines (GSEs) synthesize conversational answers from multiple sources, weakening the long-standing link between search ranking and digital visibility. This shift raises a central question for content creators: How can we reliably quantify a source article's influence on a GSE's synthesized answer across diverse intents and follow-up questions? We introduce CC-GSEO-Bench, a content-centric benchmark that couples a large-scale dataset with a creator-centered evaluation framework. The dataset contains over 1,000 source articles and over 5,000 query-article pairs, organized in a one-to-many structure for article-level evaluation. We ground construction in realistic retrieval by combining seed queries from public QA datasets with limited synthesized augmentation and retaining only queries whose paired source reappears in a follow-up retrieval step. On top of this dataset, we operationalize influence along three core dimensions: Exposure, Faithful Credit, and Causal Impact, and two content-quality dimensions: Readability and Structure, and Trustworthiness and Safety. We aggregate query-level signals over each article's query cluster to summarize influence strength, coverage, and stability, and empirically characterize influence dynamics across representative content patterns.
| 2025-12-26
| 2025-12-29
|
[
"cs.CL"
] |
Qiyuan Chen, Jiahe Chen, Hongsen Huang, Qian Shao, Jintai Chen, Renjie Hua, Hongxia Xu, Ruijia Wu, Ren Chuan, Jian Wu
|
2512.08101
|
Branching Fixed Effects: A Proposal for Communicating Uncertainty
|
Economists often rely on estimates of linear fixed effects models produced by other teams of researchers. Assessing the uncertainty in these estimates can be challenging. I propose a form of sample splitting for networks that partitions the data into statistically independent branches, each of which can be used to compute an unbiased estimate of the parameters of interest in two-way fixed effects models. These branches facilitate uncertainty quantification, moment estimation, and shrinkage. Drawing on results from the graph theory literature on tree packing, I develop algorithms to efficiently extract branches from large networks. I illustrate these techniques using a benchmark dataset from Veneto, Italy that has been widely used to study firm wage effects.
| 2025-12-26
| 2025-12-29
|
[
"econ.EM",
"stat.AP",
"stat.CO"
] |
Patrick Kline
|
2512.22091
|
Factoriality and birational rigidity of two families of singular quartic three-folds
|
In this paper we study two families of three-dimensional quartics in the complex projective space ${\mathbb P}^4$: hypersurfaces with a unique quadratic singularity of rank 3, which is resolved by two blowups, and hypersurfaces with two quadratic singularities of rank 3 and 4, respectively. Both families have codimension 3 in the natural parameter space. For a Zariski general quartic in each of these families we prove factoriality and birational rigidity and describe its group of birational self-maps.
| 2025-12-26
| 2025-12-29
|
[
"math.AG"
] |
Aleksandr V. Pukhlikov
|
2512.22317
|
LangPrecip: Language-Aware Multimodal Precipitation Nowcasting
|
Short-term precipitation nowcasting is an inherently uncertain and under-constrained spatiotemporal forecasting problem, especially for rapidly evolving and extreme weather events. Existing generative approaches rely primarily on visual conditioning, leaving future motion weakly constrained and ambiguous. We propose a language-aware multimodal nowcasting framework(LangPrecip) that treats meteorological text as a semantic motion constraint on precipitation evolution. By formulating nowcasting as a semantically constrained trajectory generation problem under the Rectified Flow paradigm, our method enables efficient and physically consistent integration of textual and radar information in latent space.We further introduce LangPrecip-160k, a large-scale multimodal dataset with 160k paired radar sequences and motion descriptions. Experiments on Swedish and MRMS datasets show consistent improvements over state-of-the-art methods, achieving over 60 \% and 19\% gains in heavy-rainfall CSI at an 80-minute lead time.
| 2025-12-26
| 2025-12-30
|
[
"cs.LG",
"cs.AI",
"cs.CV"
] |
Xudong Ling, Tianxi Huang, Qian Dong, Tao He, Chaorong Li, Guiduo Duan
|
2501.18581
|
Bias-variance decompositions: the exclusive privilege of Bregman divergences
|
Bias-variance decompositions are widely used to understand the generalization performance of machine learning models. While the squared error loss permits a straightforward decomposition, other loss functions - such as zero-one loss or $L_1$ loss - either fail to sum bias and variance to the expected loss or rely on definitions that lack the essential properties of meaningful bias and variance. Recent research has shown that clean decompositions can be achieved for the broader class of Bregman divergences, with the cross-entropy loss as a special case. However, the necessary and sufficient conditions for these decompositions remain an open question.
In this paper, we address this question by studying continuous, nonnegative loss functions that satisfy the identity of indiscernibles (zero loss if and only if the two arguments are identical), under mild regularity conditions. We prove that so-called $g$-Bregman or rho-tau divergences are the only such loss functions that have a clean bias-variance decomposition. A $g$-Bregman divergence can be transformed into a standard Bregman divergence through an invertible change of variables. This makes the squared Mahalanobis distance, up to such a variable transformation, the only symmetric loss function with a clean bias-variance decomposition. Consequently, common metrics such as $0$-$1$ and $L_1$ losses cannot admit a clean bias-variance decomposition, explaining why previous attempts have failed. We also examine the impact of relaxing the restrictions on the loss functions and how this affects our results.
| 2025-12-26
| 2025-12-29
|
[
"cs.LG"
] |
Tom Heskes
|
2512.22387
|
AI-Generated Code Is Not Reproducible (Yet): An Empirical Study of Dependency Gaps in LLM-Based Coding Agents
|
The rise of Large Language Models (LLMs) as coding agents promises to accelerate software development, but their impact on generated code reproducibility remains largely unexplored. This paper presents an empirical study investigating whether LLM-generated code can be executed successfully in a clean environment with only OS packages and using only the dependencies that the model specifies. We evaluate three state-of-the-art LLM coding agents (Claude Code, OpenAI Codex, and Gemini) across 300 projects generated from 100 standardized prompts in Python, JavaScript, and Java. We introduce a three-layer dependency framework (distinguishing between claimed, working, and runtime dependencies) to quantify execution reproducibility. Our results show that only 68.3% of projects execute out-of-the-box, with substantial variation across languages (Python 89.2%, Java 44.0%). We also find a 13.5 times average expansion from declared to actual runtime dependencies, revealing significant hidden dependencies.
| 2025-12-26
| 2025-12-30
|
[
"cs.SE",
"cs.AI",
"cs.MA"
] |
Bhanu Prakash Vangala, Ali Adibifar, Tanu Malik, Ashish Gehani
|
2511.07920
|
Lightweight Diffusion-based Framework for Online Imagined Speech Decoding in Aphasia
|
Individuals with aphasia experience severe difficulty in real-time verbal communication, while most imagined speech decoding approaches remain limited to offline analysis or computationally demanding models. To address this limitation, we propose a two-session experimental framework consisting of an offline data acquisition phase and a subsequent online feedback phase for real-time imagined speech decoding. The paradigm employed a four-class Korean-language task, including three imagined speech targets selected according to the participant's daily communicative needs and a resting-state condition, and was evaluated in a single individual with chronic anomic aphasia. Within this framework, we introduce a lightweight diffusion-based neural decoding model explicitly optimized for real-time inference, achieved through architectural simplifications such as dimensionality reduction, temporal kernel optimization, group normalization with regularization, and dual early-stopping criteria. In real-time evaluation, the proposed system achieved 65 percent top-1 and 70 percent top-2 accuracy, with the Water class reaching 80 percent top-1 and 100 percent top-2 accuracy. These results demonstrate that real-time-optimized diffusion-based architectures, combined with clinically grounded task design, can support feasible online imagined speech decoding for communication-oriented BCI applications in aphasia.
| 2025-12-26
| 2025-12-29
|
[
"cs.AI"
] |
Eunyeong Ko, Soowon Kim, Ha-Na Jo
|
2512.21900
|
Nucleon momentum distributions of complex nuclei from inclusive electron scattering
|
Nucleon momentum distributions (NMDs) reveal essential information about Fermi motion and short-range correlations (SRCs). In extracting NMDs from inclusive electron scattering data, theoretical analyses, such as the scaling analysis, are typically employed. For complex nuclei, consistently treating the excitation energy of the residual system is a complicated task, leading to discrepancies between existing extracted NMDs and ab initio calculations, particularly around the Fermi momentum $k_F$. To address this issue, we introduce an improved description of the excitation energy in the framework of the relativistic Fermi gas (RFG) model. With this treatment, the extracted NMDs of complex nuclei show better agreement with ab initio calculations across the low- and high-momentum range, especially around $k_F$, successfully reproducing both the behaviors of Fermi motion and SRCs. These results provide a new experimental perspective on the interplay between Fermi motion and SRCs in complex nuclei.
| 2025-12-26
| 2025-12-29
|
[
"nucl-th"
] |
Tongqi Liang, Dong Bai, Zhongzhou Ren
|
2512.21921
|
AutoPP: Towards Automated Product Poster Generation and Optimization
|
Product posters blend striking visuals with informative text to highlight the product and capture customer attention. However, crafting appealing posters and manually optimizing them based on online performance is laborious and resource-consuming. To address this, we introduce AutoPP, an automated pipeline for product poster generation and optimization that eliminates the need for human intervention. Specifically, the generator, relying solely on basic product information, first uses a unified design module to integrate the three key elements of a poster (background, text, and layout) into a cohesive output. Then, an element rendering module encodes these elements into condition tokens, efficiently and controllably generating the product poster. Based on the generated poster, the optimizer enhances its Click-Through Rate (CTR) by leveraging online feedback. It systematically replaces elements to gather fine-grained CTR comparisons and utilizes Isolated Direct Preference Optimization (IDPO) to attribute CTR gains to isolated elements. Our work is supported by AutoPP1M, the largest dataset specifically designed for product poster generation and optimization, which contains one million high-quality posters and feedback collected from over one million users. Experiments demonstrate that AutoPP achieves state-of-the-art results in both offline and online settings. Our code and dataset are publicly available at: https://github.com/JD-GenX/AutoPP
| 2025-12-26
| 2025-12-29
|
[
"cs.CV",
"cs.IR",
"cs.LG"
] |
Jiahao Fan, Yuxin Qin, Wei Feng, Yanyin Chen, Yaoyu Li, Ao Ma, Yixiu Li, Li Zhuang, Haoyi Bian, Zheng Zhang, Jingjing Lv, Junjie Shen, Ching Law
|
2509.12759
|
A-TDOM: Active TDOM via On-the-Fly 3DGS
|
True Digital Orthophoto Map (TDOM), a 2D objective representation of the Earth's surface, is an essential geospatial product widely used in urban management, city planning, land surveying, and related applications. However, traditional TDOM generation typically relies on a complex offline photogrammetric pipeline, leading to substantial latency and making it unsuitable for time-critical or real-time scenarios. Moreover, the quality of TDOM may deteriorate due to inaccurate camera poses, imperfect Digital Surface Model (DSM), and incorrect occlusions detection. To address these challenges, this work introduces A-TDOM, a near real-time TDOM generation method built upon On-the-Fly 3DGS (3D Gaussian Splatting) optimization. As each incoming image arrives, its pose and sparse point cloud are computed via On-the-Fly SfM. Newly observed regions are then incrementally reconstructed as additional 3D Gaussians are inserted using a Delaunay triangulated Gaussian sampling and integration and are further optimized via adaptive training iterations and learning rate, especially in previously unseen or coarsely modeled areas. With orthogonal splatting integrated into the rendering pipeline, A-TDOM can actively produce updated TDOM outputs immediately after each 3DGS update. Code is now available at https://github.com/xywjohn/A-TDOM.
| 2025-12-26
| 2025-12-29
|
[
"cs.CV"
] |
Yiwei Xu, Xiang Wang, Yifei Yu, Wentian Gan, Luca Morelli, Giulio Perda, Xin Wang, Zongqian Zhan, Fabio Remondino
|
2512.19442
|
Real-Time Streamable Generative Speech Restoration with Flow Matching
|
Diffusion-based generative models have greatly impacted the speech processing field in recent years, exhibiting high speech naturalness and spawning a new research direction. Their application in real-time communication is, however, still lagging behind due to their computation-heavy nature involving multiple calls of large DNNs.
Here, we present Stream$.$FM, a frame-causal flow-based generative model with an algorithmic latency of 32 milliseconds (ms) and a total latency of 48 ms, paving the way for generative speech processing in real-time communication. We propose a buffered streaming inference scheme and an optimized DNN architecture, show how learned few-step numerical solvers can boost output quality at a fixed compute budget, explore model weight compression to find favorable points along a compute/quality tradeoff, and contribute a model variant with 24 ms total latency for the speech enhancement task.
Our work looks beyond theoretical latencies, showing that high-quality streaming generative speech processing can be realized on consumer GPUs available today. Stream$.$FM can solve a variety of speech processing tasks in a streaming fashion: speech enhancement, dereverberation, codec post-filtering, bandwidth extension, STFT phase retrieval, and Mel vocoding. As we verify through comprehensive evaluations and a MUSHRA listening test, Stream$.$FM establishes a state-of-the-art for generative streaming speech restoration, exhibits only a reasonable reduction in quality compared to a non-streaming variant, and outperforms our recent work (Diffusion Buffer) on generative streaming speech enhancement while operating at a lower latency.
| 2025-12-26
| 2025-12-29
|
[
"eess.SP",
"cs.LG",
"cs.SD"
] |
Simon Welker, Bunlong Lay, Maris Hillemann, Tal Peer, Timo Gerkmann
|
2512.04192
|
Two dimensional de-Sitter and deformed CFTs
|
We present an alternative dimensional reduction that yields an effective theory of dilatons in a two-dimensional de Sitter background. Specifically, by performing an S-wave reduction of higher-dimensional Einstein gravity, we obtain free massless dilatons in the Nariai static patch, and a dynamically evolving dilatons in the past Milne wedge. We then propose a (Nariai) static patch worldsheet formulation in terms of CFTs with SL(2,$\mathbb{R}$) deformed Hamiltonians on the cylinder. A key feature of this construction is that a stretched horizon in the (Nariai) static patch, equipped with an emergent UV boundary condition, acts as a gravitating observer. Using the similar reduction, we have also obtained a Schwarzian action coupled to free massless dilatons in the near horizon near extremal limit of four dimensional charged AdS black holes. The worldsheet description for the same has been proposed and discussed in \cite{Das:2025cuq}. We also comment on how different notions of worldsheet time may themselves be \textit{emergent}.
| 2025-12-26
| 2025-12-29
|
[
"hep-th"
] |
Suchetan Das
|
2512.21985
|
LVLM-Aided Alignment of Task-Specific Vision Models
|
In high-stakes domains, small task-specific vision models are crucial due to their low computational requirements and the availability of numerous methods to explain their results. However, these explanations often reveal that the models do not align well with human domain knowledge, relying instead on spurious correlations. This might result in brittle behavior once deployed in the real-world. To address this issue, we introduce a novel and efficient method for aligning small task-specific vision models with human domain knowledge by leveraging the generalization capabilities of a Large Vision Language Model (LVLM). Our LVLM-Aided Visual Alignment (LVLM-VA) method provides a bidirectional interface that translates model behavior into natural language and maps human class-level specifications to image-level critiques, enabling effective interaction between domain experts and the model. Our method demonstrates substantial improvement in aligning model behavior with human specifications, as validated on both synthetic and real-world datasets. We show that it effectively reduces the model's dependence on spurious features and on group-specific biases, without requiring fine-grained feedback.
| 2025-12-26
| 2025-12-29
|
[
"cs.CV",
"cs.AI"
] |
Alexander Koebler, Lukas Kuhn, Ingo Thon, Florian Buettner
|
2512.12291
|
NICER Magnetar Burst Catalog
|
In this paper, we present a comprehensive catalog of short bursts from magnetars based on eight years of NICER observations. A total of 1130 bursts were identified, making this the largest magnetar burst catalog to date. The sample is dominated by SGR 1935+2154, which contributes 76% of all detected bursts. We analyzed burst durations, spectral properties, and their correlations across multiple sources. Bursts from SGR 1935+2154 exhibit significantly longer durations, with a mean of 317 ms, compared to a mean of 23 ms for bursts from other magnetars. Two microsecond-scale bursts were detected for the first time, originating from 1E 1048.1-5937 and CXOU J010043.1-721134. Spectral analysis in the 0.5--8 keV range using both blackbody and power-law models shows that bursts with higher fluences have harder spectra. In contrast, correlations between burst duration and spectral parameters are weak or absent. This catalog provides a valuable dataset for studying magnetar short bursts, enabling future modeling efforts and improving our understanding of the diversity and physical mechanisms of magnetar bursts.
| 2025-12-26
| 2025-12-29
|
[
"astro-ph.HE"
] |
Che-Yen Chu, Chin-Ping Hu, Teruaki Enoto, George A. Younes, Andrea Sanna, Sebastien Guillot, Rachael Stewart, Zaven Arzoumanian, Matthew G. Baring, Marlon L. Bause, Tolga Güver, Wynn C. G. Ho, Chryssa Kouveliotou, Alex Van Kooten, Zorawar Wadiasingh, Keith C. Gendreau
|
2504.10410
|
Purcell-enhanced quantum adsorption
|
Cold atoms can adsorb to a surface with the emission of a single phonon when the binding energy is sufficiently small. The effects of phonon damping and adsorbent size on the adsorption rate in this quantum regime are studied using the multimode Rabi model. It is demonstrated that the adsorption rate can be either enhanced or suppressed relative to the Fermi golden rule rate, in analogy to cavity effects in the spontaneous emission rate in QED. A mesoscopic-sized adsorbent behaves as an acoustic cavity that enhances the adsorption rate when tuned to the adsorption transition frequency and suppresses the rate when detuned. This acoustic cavity effect occurs in the regime where the frequency spacing between vibrational modes exceeds the phonon linewidth.
| 2025-12-26
| 2025-12-29
|
[
"quant-ph",
"cond-mat.mes-hall"
] |
Dennis P. Clougherty
|
2512.22078
|
Notes on some categories related to that of finite totally ordered sets
|
The purpose of these notes is to collect in one place some facts on the category of finite totally ordered sets and some related categories. More specifically, we collect some results on them which will be useful for the study of iteratedly meta theories of algebra in the style of our work arXiv:1601.00301 (arXiv:1601.00301), which is a kind of higher order universal algebra.
| 2025-12-26
| 2025-12-29
|
[
"math.CT"
] |
Takuo Matsuoka
|
2512.21828
|
Contextual Biasing for LLM-Based ASR with Hotword Retrieval and Reinforcement Learning
|
Large language model (LLM)-based automatic speech recognition (ASR) has recently achieved strong performance across diverse tasks, yet contextual biasing for named entities and hotwords under large vocabularies remains challenging. In this work, we propose a scalable two-stage framework that integrates hotword retrieval with LLM-ASR adaptation. First, we extend the Global-Local Contrastive Language-Audio pre-trained model (GLCLAP) to retrieve a compact top-k set of hotword candidates from a large vocabulary via robustness-aware data augmentation and fuzzy matching. Second, we inject the retrieved candidates as textual prompts into an LLM-ASR model and fine-tune it with Generative Rejection-Based Policy Optimization (GRPO), using a task-driven reward that jointly optimizes hotword recognition and overall transcription accuracy. Experiments on hotword-focused test sets show substantial keyword error rate (KER) reductions while maintaining sentence accuracy on general ASR benchmarks, demonstrating the effectiveness of the proposed framework for large-vocabulary contextual biasing.
| 2025-12-26
| 2025-12-29
|
[
"eess.AS"
] |
YuXiang Kong, JunFeng Hou, Jian Tang, Bingqing Zhu, Jicheng Zhang, Shaofei Xue
|
2510.19130
|
Denoising Complex Covariance Matrices with Hybrid ResNet and Random Matrix Theory: Cryptocurrency Portfolio Applications
|
Covariance matrices estimated from short, noisy, and non-Gaussian financial time series are notoriously unstable. Empirical evidence suggests that such covariance structures often exhibit power-law scaling, reflecting complex, hierarchical interactions among assets. Motivated by this observation, we introduce a power-law covariance model to characterize collective market dynamics and propose a hybrid estimator that integrates Random Matrix Theory (RMT) with deep Residual Neural Networks (ResNets). The RMT component regularizes the eigenvalue spectrum in high-dimensional noisy settings, while the ResNet learns data-driven corrections that recover latent structural dependencies encoded in the eigenvectors. Monte Carlo simulations show that the proposed ResNet-based estimators consistently minimize both Frobenius and minimum-variance losses across a range of population covariance models. Empirical experiments on 89 cryptocurrencies over the period 2020-2025, using a training window ending at the local Bitcoin peak in November 2021 and testing through the subsequent bear market, demonstrate that a two-step estimator combining hierarchical filtering with ResNet corrections produces the most profitable and well-balanced portfolios, remaining robust across market regime shifts. Beyond finance, the proposed hybrid framework applies broadly to high-dimensional systems described by low-rank deformations of Wishart ensembles, where incorporating eigenvector information enables the detection of multiscale and hierarchical structure that is inaccessible to purely eigenvalue-based methods.
| 2025-12-26
| 2025-12-30
|
[
"q-fin.CP"
] |
Andres Garcia-Medina
|
2512.22382
|
Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration
|
Hyperparameter tuning can dramatically impact training stability and final performance of large-scale models. Recent works on neural network parameterisations, such as $μ$P, have enabled transfer of optimal global hyperparameters across model sizes. These works propose an empirical practice of search for optimal global base hyperparameters at a small model size, and transfer to a large size. We extend these works in two key ways. To handle scaling along most important scaling axes, we propose the Complete$^{(d)}$ Parameterisation that unifies scaling in width and depth -- using an adaptation of CompleteP -- as well as in batch-size and training duration. Secondly, with our parameterisation, we investigate per-module hyperparameter optimisation and transfer. We characterise the empirical challenges of navigating the high-dimensional hyperparameter landscape, and propose practical guidelines for tackling this optimisation problem. We demonstrate that, with the right parameterisation, hyperparameter transfer holds even in the per-module hyperparameter regime. Our study covers an extensive range of optimisation hyperparameters of modern models: learning rates, AdamW parameters, weight decay, initialisation scales, and residual block multipliers. Our experiments demonstrate significant training speed improvements in Large Language Models with the transferred per-module hyperparameters.
| 2025-12-26
| 2025-12-30
|
[
"cs.LG",
"cs.AI",
"stat.ML"
] |
Bruno Mlodozeniec, Pierre Ablin, Louis Béthune, Dan Busbridge, Michal Klein, Jason Ramapuram, Marco Cuturi
|
2512.21858
|
Professor Hideki Yukawa's Anguish and a Lifelong Decision During a Three-Day Visit to Kochi to Unveil His First Bronze Statue: From a Cave Bat to the World
|
In 1954, following a five-year research period in the U.S., Professor Hideki Yukawa returned to Japan and visited Kochi on March 21 to attend the unveiling ceremony for the first statue of him ever built in Japan, a project initiated by the PTA of Yasu Elementary School in Yasu Town, Kochi Prefecture. By a coincidence of history, just three weeks prior on March 1, the U.S. had conducted a hydrogen bomb test at Bikini Atoll in the Pacific Ocean. Many Japanese fishing boats were operating there at the time and had not been informed in advance. As a result, numerous boats, including the Daigo Fukuryu Maru, were exposed to radiation. Upon his arrival at Kochi Station on the evening of March 21, Yukawa was relentlessly questioned by reporters about the Bikini hydrogen bomb. This was a source of deep anguish for Yukawa, a Japanese physicist who had won the Nobel Prize for his work on "atomic physics." He firmly refused to answer, stating that the topic was "outside the scope of my research." The next evening, at a public lecture in Kochi City on March 22, he again refused to speak about the Bikini hydrogen bomb or nuclear power, stating that he was an amateur in nuclear research and that there were many other experts. However, just four days later, on March 28, after returning to Kyoto, Yukawa drafted his famous essay, "The Turning Point for Humanity and Atomic Power," which was published in a newspaper on March 30. From that point on, he was drawn into the tumultuous issue of the Bikini hydrogen bomb and nuclear power. When did a tormented Yukawa make his decision? This article meticulously reveals, based on historical documents, what led the anguished Yukawa to make such a rapid decision within a single day and what caused the immense change in his mindset overnight.
| 2025-12-26
| 2025-12-29
|
[
"physics.hist-ph"
] |
Shigeo Ohkubo
|
2512.22322
|
SmartSnap: Proactive Evidence Seeking for Self-Verifying Agents
|
Agentic reinforcement learning (RL) holds great promise for the development of autonomous agents under complex GUI tasks, but its scalability remains severely hampered by the verification of task completion. Existing task verification is treated as a passive, post-hoc process: a verifier (i.e., rule-based scoring script, reward or critic model, and LLM-as-a-Judge) analyzes the agent's entire interaction trajectory to determine if the agent succeeds. Such processing of verbose context that contains irrelevant, noisy history poses challenges to the verification protocols and therefore leads to prohibitive cost and low reliability. To overcome this bottleneck, we propose SmartSnap, a paradigm shift from this passive, post-hoc verification to proactive, in-situ self-verification by the agent itself. We introduce the Self-Verifying Agent, a new type of agent designed with dual missions: to not only complete a task but also to prove its accomplishment with curated snapshot evidences. Guided by our proposed 3C Principles (Completeness, Conciseness, and Creativity), the agent leverages its accessibility to the online environment to perform self-verification on a minimal, decisive set of snapshots. Such evidences are provided as the sole materials for a general LLM-as-a-Judge verifier to determine their validity and relevance. Experiments on mobile tasks across model families and scales demonstrate that our SmartSnap paradigm allows training LLM-driven agents in a scalable manner, bringing performance gains up to 26.08% and 16.66% respectively to 8B and 30B models. The synergizing between solution finding and evidence seeking facilitates the cultivation of efficient, self-verifying agents with competitive performance against DeepSeek V3.1 and Qwen3-235B-A22B.
| 2025-12-26
| 2025-12-30
|
[
"cs.CL",
"cs.AI",
"cs.CV",
"cs.LG",
"cs.MA"
] |
Shaofei Cai, Yulei Qin, Haojia Lin, Zihan Xu, Gang Li, Yuchen Shi, Zongyi Li, Yong Mao, Siqi Cai, Xiaoyu Tan, Yitao Liang, Ke Li, Xing Sun
|
2512.22056
|
Enhanced Distributed Variational Quantum Eigensolver for Large-Scale MaxCut Problem
|
MaxCut is a canonical NP-hard combinatorial optimization problem in graph theory with broad applications ranging from physics to bioinformatics. Although variational quantum algorithms offer promising new approaches that may eventually outperform classical schemes, they suffer from resource constraints and trainability issues such as barren plateaus, making large-scale instances intractable on noisy intermediate-scale quantum devices. In this paper, we propose an enhanced distributed variational quantum eigensolver for large-scale MaxCut problems, which extends our prior distributed variational quantum eigensolver framework by integrating a novel hybrid classical-quantum perturbation strategy, enhances optimization scalability and efficiency. Our algorithm solves weighted MaxCut instances with up to 1000 vertices using only 10 qubits, and numerical results indicate that it consistently outperforms the Goemans-Williamson algorithm. We further employ a warm-start initialization strategy, seeding the algorithm with high-quality solutions from the Goemans-Williamson algorithm, with results confirming that the optimal classical solution can be effectively further improved. The practical utility of the proposed algorithm is further validated through its application to haplotype phasing on genome sequencing data of the human ABCA1 gene, producing high-quality haplotypes that rival those obtained by the Goemans-Williamson algorithm with $10^6$ projections. These results establish the proposed algorithm as a scalable, NISQ-compatible framework for near-term quantum-enhanced large-scale combinatorial optimization.
| 2025-12-26
| 2025-12-29
|
[
"quant-ph"
] |
Yuefeng Lin, Kun Wang, Qinyuan Zheng, Rui Zhang, Jing-Kai Fang, Tiejun Meng, Jingen Xiang, Cong Guo, Jun-Han Huang
|
2406.06804
|
Robustness to missing data: breakdown point analysis
|
Missing data is pervasive in econometric applications, and rarely is it plausible that the data are missing (completely) at random. This paper proposes a methodology for studying the robustness of results drawn from incomplete datasets. Selection is measured as the divergence from the distribution of complete observations to the distribution of incomplete observations. The breakdown point is defined as the minimal amount of selection needed to overturn a given result. Reporting point estimates and lower confidence intervals of the breakdown point is a simple, concise way to communicate the robustness of a result. An estimator of the breakdown point is proposed and shown root-n consistent and asymptotically normal. This estimator can be applied directly to conclusions drawn from any model identified with the generalized method of moments (GMM) that satisfies mild assumptions. Simulations demonstrate the finite sample performance of the breakdown point estimator on averages, linear regression, and logistic regression. The methodology is illustrated by estimating the breakdown point of conclusions drawn from several randomized controlled trails suffering from missing data due to attrition.
| 2025-12-26
| 2025-12-29
|
[
"econ.EM"
] |
Daniel Ober-Reynolds
|
2512.22030
|
A Fundamental Theorem on Einstein-Podolsky-Rosen Steering
|
Quantum nonlocality is an essential nonlocality resource in quantum information. It has been classified into three distinct types: quantum entanglement, Einstein-Podolsky-Rosen (EPR) steering, and Bell's nonlocality. In 1991, Gisin presented a fundamental theorem on Bell's nonlocality, pointing out all pure entangled states possess Bell's nonloclaity. Many of the core protocols of quantum information science (such as quantum teleportation, quantum key distribution, and certain algorithms in quantum computing) rely on entanglement. Gisin's theorem tells us that as long as we successfully prepare a pure entangled state, we then have a Bell-nonlocality resource that can show the non-classical correlations. Such a resource is not ``virtual'' and can be tested and used through Bell-experiments. Similarly, in this work, we present a Gisin-like fundamental theorem on EPR steering, which indicates all rank-2 (and rank-1) entangled states possess EPR steerability. Thus all rank-2 entangled states can be applicable as EPR-steering resources in quantum information.
| 2025-12-26
| 2025-12-29
|
[
"quant-ph"
] |
Yu-Xuan Zhang, Jing-Ling Chen
|
2512.21862
|
Nonparametric methods for comparing distribution functionals for dependent samples with application to inequality measures
|
This paper proposes asymptotically distribution-free inference methods for comparing a broad range of welfare indices across dependent samples, including those employed in inequality, poverty, and risk analysis. Two distinct situations are considered. \emph{First}, we propose asymptotic and bootstrap intersection methods which are completely robust to arbitrary dependence between two samples. \emph{Second}, we focus on the common case of overlapping samples -- a special form of dependent samples where sample dependence arises solely from matched pairs -- and provide asymptotic and bootstrap methods for comparing indices. We derive consistent estimates for asymptotic variances using the influence function approach. The performance of the proposed methods is studied in a simulation experiment: we find that confidence intervals with overlapping samples exhibit satisfactory coverage rates with reasonable precision, whereas conventional methods based on an assumption of independent samples have an inferior performance in terms of coverage rates and interval widths. Asymptotic inference can be less reliable when dealing with heavy-tailed distributions, while the bootstrap method provides a viable remedy, unless the variance is substantial or nonexistent. The intersection method yields reliable results with arbitrary dependent samples, including instances where overlapping samples are not feasible. We demonstrate the practical applicability of our proposed methods in analyzing dynamic changes in household financial inequality in Italy over time.
| 2025-12-26
| 2025-12-29
|
[
"econ.EM"
] |
Jean-Marie Dufour, Tianyu He
|
2512.22002
|
The algebro-geometric aspect of the iterated limit of a quaternary of means of four terms
|
We study the iterated limit of a quaternary of means of four terms through the period map from the family of cyclic fourfold coverings of the complex projective line branching at six points to the three-dimensional complex ball $\mathbb{B}_3$ embedded into the Siegel upper half-space of degree four. We construct four automorphic forms on $\mathbb{B}_3$ expressing the inverse of the period map, and give an equality between one of them and a period integral, which is an analogy of Jacobi's formula between a theta constant and an elliptic integral. We find a transformation of $\mathbb{B}_3$ such that the quaternary of means appears by its actions on the four automorphic forms. These results enable us to express the iterated limit by the Lauricella hypergeometric series of type $D$ in three variables.
| 2025-12-26
| 2025-12-29
|
[
"math.AG"
] |
Keiji Matsumoto, Ryunosuke Nakano
|
2512.22368
|
Introduction to Lattice Field Theory
|
This chapter provides a pedagogical introduction to lattice quantum field theory, with strong emphasis on lattice quantum chromodynamics. The chapter reviews key foundational concepts of lattice quantum chromodynamics, as well as a broad summary of ongoing research in the field.
| 2025-12-26
| 2025-12-30
|
[
"hep-lat"
] |
Raúl A. Briceño
|
2512.21992
|
Measure of entanglement and the monogamy relation: a topical review
|
Characterizing entanglement, including quantifying and distribution of entanglement, which lies at heart of the quantum resource theory, have been investigated extensively ever since Bennett \etal proposed three seminal measures of entanglement in 1996. Up to now, there are numerous measures of entanglement that have been proposed from different point of view and plenty of monogamy relations have been explored which make the distribution of entanglement became more and more clear. While this is relatively easy in the case of pure states, it is much more intricate for the case of mixed quantum states especially with higher dimension and more particles in the system. We present here an overview of the theory along this line. We outline most of the results in this field historically and focus on the finite-dimensional systems. In particular we emphasize the point of view that (i) which yardsticks haven been applied in quantifying entanglement and its distribution, (ii) what are the substantive characteristics and interrelations of these measures and their monogamy relations mathematically by comparing, and (iii) which concepts should be improved or revised and how they were developed accordingly.
| 2025-12-26
| 2025-12-29
|
[
"quant-ph"
] |
Yu Guo, Zhixiang Jin
|
2512.21986
|
Time-integrated Optimal Transport: A Robust Minimax Framework
|
Comparing time series in a principled manner requires capturing both temporal alignment and distributional similarity of features. Optimal transport (OT) has recently emerged as a powerful tool for this task, but existing OT-based approaches often depend on manually selected balancing parameters and can be computationally intensive. In this work, we introduce the Time-integrated Optimal Transport (TiOT) framework, which integrates temporal and feature components into a unified objective and yields a well-defined metric on the space of probability measures. This metric preserves fundamental properties of the Wasserstein distance, while avoiding the need for parameter tuning. To address the corresponding computational challenges, we introduce an entropic regularized approximation of TiOT, which can be efficiently solved using a block coordinate descent algorithm. Extensive experiments on both synthetic and real-world time series datasets demonstrate that our approach achieves improved accuracy and stability while maintaining comparable efficiency.
| 2025-12-26
| 2025-12-29
|
[
"math.OC"
] |
Thai P. D. Nguyen, Hong T. M. Chu, Kim-Chuan Toh
|
2512.22018
|
Prediction intervals for quantile autoregression
|
This paper introduces new methods for constructing prediction intervals using quantile-based techniques. The procedures are developed for both classical (homoscedastic) autoregressive models and modern quantile autoregressive models. They combine quantile estimation with multiplier bootstrap schemes to approximate the sampling variability of coefficient estimates, together with bootstrap replications of future observations. We consider both percentile-based and predictive-root-based constructions. Theoretical results establish the validity and pertinence of the proposed methods. Simulation experiments evaluate their finite-sample performance and show that the proposed methods yield improved coverage properties and computational efficiency relative to existing approaches in the literature. The empirical usefulness of the methods is illustrated through applications to U.S. unemployment rate data and retail gasoline prices.
| 2025-12-26
| 2025-12-29
|
[
"stat.ME"
] |
Silvia Novo, César Sánchez-Sellero
|
2512.22037
|
Sharp pointwise convergence of Schrödinger mean with complex time in higher dimensions
|
In this paper, we establish the almost everywhere convergence of solutions to the Schrödinger operator with complex time $ P_γf(x,t) $ in higher dimensions, under the assumption that the initial data $f$ belongs to the Sobolev space $ H^{s}(\mathbb{R}^d)$.
| 2025-12-26
| 2025-12-29
|
[
"math.AP"
] |
Meng Wang, Zhichao Wang
|
2512.22014
|
HWL-HIN: A Hypergraph-Level Hypergraph Isomorphism Network as Powerful as the Hypergraph Weisfeiler-Lehman Test with Application to Higher-Order Network Robustness
|
Robustness in complex systems is of significant engineering and economic importance. However, conventional attack-based a posteriori robustness assessments incur prohibitive computational overhead. Recently, deep learning methods, such as Convolutional Neural Networks (CNNs) and Graph Neural Networks (GNNs), have been widely employed as surrogates for rapid robustness prediction. Nevertheless, these methods neglect the complex higher-order correlations prevalent in real-world systems, which are naturally modeled as hypergraphs. Although Hypergraph Neural Networks (HGNNs) have been widely adopted for hypergraph learning, their topological expressive power has not yet reached the theoretical upper bound. To address this limitation, inspired by Graph Isomorphism Networks, this paper proposes a hypergraph-level Hypergraph Isomorphism Network framework. Theoretically, this approach is proven to possess an expressive power strictly equivalent to the Hypergraph Weisfeiler-Lehman test and is applied to predict hypergraph robustness. Experimental results demonstrate that while maintaining superior efficiency in training and prediction, the proposed method not only outperforms existing graph-based models but also significantly surpasses conventional HGNNs in tasks that prioritize topological structure representation.
| 2025-12-26
| 2025-12-29
|
[
"cs.LG"
] |
Chengyu Tian, Wenbin Pei
|
2512.22302
|
Statistical and Machine Learning Analysis of Traffic Accidents on US 158 in Currituck County: A Comparison with HSM Predictions
|
This study extends previous hotspot and Chi-Square analysis by Sawyer \cite{sawyer2025hotspot} by integrating advanced statistical analysis, machine learning, and spatial modeling techniques to analyze five years (2019--2023) of traffic accident data from an 8.4-mile stretch of US 158 in Currituck County, NC. Building upon foundational statistical work, we apply Kernel Density Estimation (KDE), Negative Binomial Regression, Random Forest classification, and Highway Safety Manual (HSM) Safety Performance Function (SPF) comparisons to identify comprehensive temporal and spatial crash patterns. A Random Forest classifier predicts injury severity with 67\% accuracy, outperforming HSM SPF. Spatial clustering is confirmed via Moran's I test ($I = 0.32$, $p < 0.001$), and KDE analysis reveals hotspots near major intersections, validating and extending earlier hotspot identification methods. These results support targeted interventions to improve traffic safety on this vital transportation corridor. Our objective is to provide actionable insights for improving safety on US 158 while contributing to the broader understanding of rural highway safety analysis through methodological advancement beyond basic statistical techniques.
| 2025-12-26
| 2025-12-30
|
[
"cs.LG"
] |
Jennifer Sawyer, Julian Allagan
|
2512.22017
|
Necessary conditions for spin-resolved Josephson diode effect across strongly spin-polarized magnetic materials
|
We present a set of necessary conditions for the appearance of charge and spin Josephson diode effects across strongly spin-polarized inhomogeneous magnetic materials (FM) placed between two spin-singlet superconductors. Noncoplanarity of the FM's spin texture gives rise to quantum geometric phases, $ÎÏ'$, that enter the Josephson current-phase relation (CPR) similarly to the superconducting phase difference, resulting in charge and spin Josephson diode effects. Our study shows that such effects appear if the CPR possesses no phase-inversion center, achieved under the following conditions. First, noncoplanarity of the spin texture is necessary to break the spatial inversion symmetry. Second, both spin bands have to contribute to the transport, i.e., the effect is absent in half-metallic junctions. Third, different band-specific densities of states are required, and this condition is ensured by the strong spin polarization of the FM. Finally, higher harmonics in the CPR are necessary, i.e., the effect is absent in the tunneling limit. However, even in this case, the CPR must not have a phase-inversion center, which is ensured by the restriction of the quantum geometric phase to values $ÎÏ'\neq kÏ/2, k\in\mathbb{Z}$. We formulate a minimal phenomenological model that incorporates all these points, qualitatively illustrating our theory.
| 2025-12-26
| 2025-12-29
|
[
"cond-mat.supr-con",
"cond-mat.mes-hall"
] |
Danilo NikoliÄ, Niklas L. Schulz, Matthias Eschrig
|
2512.22115
|
Long time dynamics of space periodic water waves
|
We review recent advances regarding the long-time dynamics of space-periodic water waves, focusing on 1) bifurcation of quasi-periodic solutions, both standing and traveling; 2) long-time well-posedness results; 3) modulational instability of Stokes waves. These results rely on unconventional approaches to KAM and Birkhoff normal form theories for Hamiltonian quasi-linear PDEs and symplectic Kato perturbation theory for separated eigenvalues of reversible and Hamiltonian operators.
| 2025-12-26
| 2025-12-29
|
[
"math.AP"
] |
Massimiliano Berti
|
2412.09614
|
RAVEL: Rare Concept Generation and Editing via Graph-driven Relational Guidance
|
Despite impressive visual fidelity, current text-to-image (T2I) diffusion models struggle to depict rare, complex, or culturally nuanced concepts due to training data limitations. We introduce RAVEL, a training-free framework that significantly improves rare concept generation, context-driven image editing, and self-correction by integrating graph-based retrieval-augmented generation (RAG) into diffusion pipelines. Unlike prior RAG and LLM-enhanced methods reliant on visual exemplars, static captions or pre-trained knowledge of models, RAVEL leverages structured knowledge graphs to retrieve compositional, symbolic, and relational context, enabling nuanced grounding even in the absence of visual priors. To further refine generation quality, we propose SRD, a novel self-correction module that iteratively updates prompts via multi-aspect alignment feedback, enhancing attribute accuracy, narrative coherence, and semantic fidelity. Our framework is model-agnostic and compatible with leading diffusion models including Stable Diffusion XL, Flux, and DALL-E 3. We conduct extensive evaluations across three newly proposed benchmarks - MythoBench, Rare-Concept-1K, and NovelBench. RAVEL also consistently outperforms SOTA methods across perceptual, alignment, and LLM-as-a-Judge metrics. These results position RAVEL as a robust paradigm for controllable and interpretable T2I generation in long-tail domains.
| 2025-12-26
| 2025-12-30
|
[
"cs.CV",
"cs.CL"
] |
Kavana Venkatesh, Yusuf Dalva, Ismini Lourentzou, Pinar Yanardag
|
2512.21853
|
MoonBot: Modular and On-Demand Reconfigurable Robot Toward Moon Base Construction
|
The allure of lunar surface exploration and development has recently captured widespread global attention. Robots have proved to be indispensable for exploring uncharted terrains, uncovering and leveraging local resources, and facilitating the construction of future human habitats. In this article, we introduce the modular and on-demand reconfigurable robot (MoonBot), a modular and reconfigurable robotic system engineered to maximize functionality while operating within the stringent mass constraints of lunar payloads and adapting to varying environmental conditions and task requirements. This article details the design and development of MoonBot and presents a preliminary field demonstration that validates the proof of concept through the execution of milestone tasks simulating the establishment of lunar infrastructure. These tasks include essential civil engineering operations, infrastructural component transportation and deployment, and assistive operations with inflatable modules. Furthermore, we systematically summarize the lessons learned during testing, focusing on the connector design and providing valuable insights for the advancement of modular robotic systems in future lunar missions.
| 2025-12-26
| 2025-12-29
|
[
"cs.RO",
"cs.AI"
] |
Kentaro Uno, Elian Neppel, Gustavo H. Diaz, Ashutosh Mishra, Shamistan Karimov, A. Sejal Jain, Ayesha Habib, Pascal Pama, Hazal Gozbasi, Shreya Santra, Kazuya Yoshida
|
2512.21864
|
The trinacria graphs $T_{(b+2)b2}$ are $e$-positive
|
In this paper, we identify a new family of $e$-positive graphs, called the trinacria graphs $T_{(b+2)b2}$, thereby providing a partial answer to Stanley's question on which graphs are $e$-positive. The trinacria graph $T_{abc}$ is the graph on $a+b+c+3$ vertices obtained by attaching paths $P_a$, $P_b$ and~$P_c$ to the vertices of a triangle, respectively. Our proof relies on several ad hoc combinatorial ideas, and employs divide-and-conquer techniques, charging arguments, and progressive repair methods.
| 2025-12-26
| 2025-12-29
|
[
"math.CO"
] |
Simon Y. M. Gong, David G. L. Wang, K. Zhang
|
2512.21958
|
Flow morphology and patterns in porous media convection: A persistent homology analysis
|
Convective mixing in porous media is crucial in both geophysical and industrial fields, spanning applications ranging from carbon dioxide sequestration to contaminant transport in groundwater. Key processes are affected by convective heat transport or diffusion of chemical species in porous formations. Intense convection flow and mixing create complex, dynamic patterns that are difficult to predict and measure. The present work focuses on the use of topological data analysis, in particular, the measures emerging from the growing field of persistent homology (PH), to quantify these patterns. These measures are objective and quantify structures across all temperature or concentration values simultaneously. These techniques, when applied to classical porous media setups, such as one-sided and Rayleigh-Bénard flow configurations, provide new insights into the system's structure, flow patterns, and macroscopic mixing properties. Using large datasets we make publicly available, comprising original simulations as well as those presented in previous works, we correlate the behaviour of the heat transport rate (quantified by the Nusselt number) with the evolution of the flow structures (quantified by the PH measures). Finally, we provide a detailed analysis of the flow evolution over a wide range of governing parameters, namely the Rayleigh-Darcy number and the domain size.
| 2025-12-26
| 2025-12-29
|
[
"physics.flu-dyn",
"physics.geo-ph"
] |
Marco De Paoli, Sergio Pirozzoli, Lou Kondic
|
2512.21895
|
Drug discovery guided by maximum drug likeness
|
To overcome the high attrition rate and limited clinical translatability in drug discovery, we introduce the concept of Maximum Drug-Likeness (MDL) and develop an applicable Fivefold MDL strategy (5F-MDL) to reshape the screening paradigm. The 5F-MDL strategy integrates an ensemble of 33 deep learning sub-models to construct a 33-dimensional property spectrum that quantifies the global phenotypic alignment of candidate molecules with clinically approved drugs along five axes: physicochemical properties, pharmacokinetics, efficacy, safety, and stability. Using drug-likeness scores derived from this 33-dimensional profile, we prioritized 15 high-potential molecules from a 16-million-molecule library. Experimental validation demonstrated that the lead compound M2 not only exhibits potent antibacterial activity, with a minimum inhibitory concentration (MIC) of 25.6 ug/mL, but also achieves binding stability superior to cefuroxime, as indicated by Molecular Mechanics Poisson-Boltzmann surface area (MM-PBSA) calculations of -38.54 kcal/mol and a root-mean-square deviation (RMSD) of 2.8 A. This strategy could overcome scaffold constraints and offers an efficient route for discovering lead compounds with favorable prospects against drug-resistant bacteria.
| 2025-12-26
| 2025-12-29
|
[
"q-bio.QM"
] |
Hao-Yu Zhu, Shi-Jie Du, Lu Xu, Wei Shi
|
2502.12765
|
Approximation results for weak solutions of stochastic partial differential equations
|
In probability theory, how to approximate the solution of a stochastic differential equation is an important topic. In Watanabe's classical textbook, by an approximation of the Wiener process, solutions of approximated equations converge to the solution of the stochastic differential equation in probability. In traditional approximation theorems, solutions do not contain the spatial variable. In recent years, stochastic partial differential equations have been playing major roles in probability theory. If the solution is a weak one with the spatial variable, we may not be able to directly apply these classical approximation results. In this work, we try to extend the approximation result to stochastic partial differential equations case. We show that in this case, the approximation result still holds.
| 2025-12-26
| 2025-12-29
|
[
"math.PR",
"math.AP"
] |
Xi Lin
|
2512.21854
|
Necking of epithelial tissues with cellular topological transition
|
As the cover of embryos and adult organisms, epithelial tissues are subjected to substantial mechanical forces in tissue morphogenesis. However, the finite deformation behaviors of epithelial tissues remain largely unexplored. This study combines discrete vertex simulations with a multiscale constitutive model to investigate the necking behavior of epithelial tissues. In the multiscale model, the shape changes and topological transitions of single cells are mapped to the elastic and inelastic tissue deformations via a mean-field formulation. Our results show that the necking bifurcation of a stretched tissue arises from cellular topological transitions. The bifurcation condition and the steady state of necking propagation are predicted from the constitutive model and validated by vertex simulations. Furthermore, we find that topological defects in disordered tissues facilitate necking bifurcation but impede its propagation. These defects also induce the necked region to collapse into a thin thread, as observed in real tissues. Together, our work provides valuable insights into the deformation behaviors of epithelial tissues.
| 2025-12-26
| 2025-12-29
|
[
"physics.bio-ph"
] |
Yuan He, Shi-Lei Xue
|
2512.22361
|
Traversable ghost wormholes
|
Ghost stars are compact configurations characterized by an arbitrarily small total mass. Such objects require regions of negative energy density -a condition typically regarded as unphysical within the context of conventional stellar models. Nevertheless, negative energy densities arise naturally in traversable wormhole geometries, where the violation of the null energy condition is essential to sustain the flaring-out behavior at the throat. This connection suggests that ghost-like configurations may find a natural realization within wormhole physics. In this work, we investigate the existence of ghost configurations by analyzing their associated Hawking mass. Although in spherical symmetry the Misner and Hawking masses are known to coincide, we show that when the ghost condition is extended beyond spherical symmetry and applied to the Hawking mass, it faces topological obstructions that hinder its straightforward realization. As a concrete example, we demonstrate that a Casimir-like traversable wormhole can be naturally constructed within this framework. Finally, to illustrate the properties of the resulting geometry, we analyze its Penrose-Carter diagram.
| 2025-12-26
| 2026-01-01
|
[
"gr-qc"
] |
Alberto Guilabert, Ernesto Fuenmayor, Pedro Bargueño, Ernesto Contreras
|
2512.22108
|
Keffer-like form of the symmetric Heisenberg exchange integral: Contribution to the Landau--Lifshitz--Gilbert equation and spin wave dispersion dependence
|
The symmetric Heisenberg exchange interaction and antisymmetric Dzyaloshinskii-Moriya interaction are parts of the tensor potential describing effective spin-spin interaction caused by the superexchange interaction of magnetic ions via nonmagnetic ion. There is the Keffer form of the vector constant of the Dzyaloshinskii-Moriya interaction, which includes the shift of the nonmagnetic ion (ligand) from the line connecting two magnetic ions. It is suggested, in this paper, that the ligand shift can give contribution in the constant of the symmetric Heisenberg interaction in antiferromagnetic or ferrimagnetic materials. Hence, the constant of the Heisenberg interaction is composed minimum of two terms. One does not depend on the ligand shift an gives standard contribution in the energy density like term with no derivatives of the spin densities or term containing two spatial derivatives of the spin densities. It is demonstrated that additional term gives a term in the energy density containing one spatial derivative of the spin density. Corresponding contribution in the Landau--Lifshitz--Gilbert equation is found. Possibility of the noncollinear equilibrium order of spin under influence of new spin torque is discussed. Modification of the spin wave (normal modes) dispersion dependencies in the antiferromagnetic materials is found for the collinear order and for the cycloidal order of spins. Effective spin current is derived and applied for the spin-current model of the polarization origin in multiferroics.
| 2025-12-26
| 2025-12-29
|
[
"cond-mat.mtrl-sci"
] |
Pavel A. Andreev
|
2508.02619
|
Impact of Non-Thermal Leptogenesis with Early Matter Domination on Gravitational Waves from First-order Phase Transition
|
We study the impact of non-thermal leptogenesis on the spectrum of gravitational waves (GWs) produced by a strong first-order phase transition in the early Universe. We consider a scenario in which a heavy scalar field, $Ï$, dominates the energy density of the early Universe and decays into heavy right-handed neutrinos (RHNs). The subsequent decay of RHNs generates a lepton asymmetry, which is partially converted into the observed baryon asymmetry via the sphaleron process. The $Ï$-dominated era and the entropy injection from the decays of $Ï$ and RHNs leave characteristic imprints on the GW spectrum, such as damping and modified frequency dependence, that distinguish it from the standard cosmological evolution. We identify the parameter space in which non-thermal leptogenesis is successful, leading to distinctive GW spectral features. We show that these GW signals can fall within the sensitivity ranges of future detectors such as ET, DECIGO and BBO. If observed, they would provide valuable insights into the thermal history and dynamics of the early Universe.
| 2025-12-26
| 2025-12-29
|
[
"hep-ph",
"astro-ph.CO"
] |
Dilip Kumar Ghosh, Anish Ghoshal, Koustav Mukherjee, Nimmala Narendra, Nobuchika Okada
|
2512.20458
|
Laser: Governing Long-Horizon Agentic Search via Structured Protocol and Context Register
|
Recent advances in Large Language Models (LLMs) and Large Reasoning Models (LRMs) have enabled agentic search systems that interleave multi-step reasoning with external tool use. However, existing frameworks largely rely on unstructured natural-language reasoning and accumulate raw intermediate traces in the context, which often leads to unstable reasoning trajectories, context overflow, and degraded performance on complex multi-hop queries. In this study, we introduce Laser, a general framework for stabilizing and scaling agentic search. Laser defines a symbolic action protocol that organizes agent behaviors into three spaces: planning, task-solving, and retrospection. Each action is specified with explicit semantics and a deterministic execution format, enabling structured and logical reasoning processes and reliable action parsing. This design makes intermediate decisions interpretable and traceable, enhancing explicit retrospection and fine-grained control over reasoning trajectories. In coordination with parsable actions, Laser further maintains a compact context register that stores only essential states of the reasoning process, allowing the agent to reason over long horizons without uncontrolled context expansion. Experiments on Qwen2.5/3-series models across challenging multi-hop QA datasets show that Laser consistently outperforms existing agentic search baselines under both prompting-only and fine-tuning settings, demonstrating that Laser provides a principled and effective foundation for robust, scalable agentic search.
| 2025-12-26
| 2025-12-29
|
[
"cs.IR"
] |
Shuting Wang, Qiaolin Xia, Vich Wang, Herberttli, Bobsimons, Zhicheng Dou
|
2601.02394
|
Hydrodynamic Whispering: Enabling Near-Field Silent Communication via Artificial Lateral Line Arrays
|
To address the imperative for covert underwater swarm coordination, this paper introduces "Hydrodynamic Whispering," a near-field silent communication paradigm utilizing Artificial Lateral Line (ALL) arrays. Grounded in potential flow theory, we model the transmitter as an oscillating dipole source. The resulting pressure field exhibits steep nearfield attenuation (scaling with 1/r^2, naturally delimiting a secure "communication bubble" with intrinsic Low Probability of Interception (LPI) properties. We propose a transceiver architecture featuring a Binary Phase Shift Keying (BPSK) modulation scheme adapted for mechanical actuator inertia, coupled with a bio-inspired 24-sensor conformal array. To mitigate low Signal-to-Noise Ratio (SNR) in turbulent environments,a Spatio-Temporal Joint Processing framework incorporating Spatial Matched-Field Beamforming is developed. Simulation results demonstrate that the system achieves an array gain of approximately 13.8 dB and maintains a near-zero Bit Error Rate (BER) within the effective range. This study validates the feasibility of utilizing localized hydrodynamic pressure fluctuations for reliable and secure short-range underwater networking.
| 2025-12-26
| 2026-01-07
|
[
"eess.SP",
"physics.flu-dyn"
] |
Yuan-Jie Chen
|
2512.21891
|
Non-polynomial divided difference and blossoming
|
Two notable examples of dual functionals in approximation theory and computer-aided geometric design are the blossom and the divided difference operator. Both of these dual functionals satisfy a similar set of formulas and identities. Moreover, the divided differences of polynomials can be expressed in terms of the blossom. In this paper, an extended non-polynomial homogeneous blossom for a wide collection of spline spaces, including trigonometric splines, hyperbolic splines, and special Müntz spaces of splines, is defined. It is shown that there is a relation between the non-polynomial divided difference and the blossom, which is analogous to the polynomial case.
| 2025-12-26
| 2025-12-29
|
[
"math.NA",
"cs.NA"
] |
Fatma Zürnacı-YetiÅ
|
2501.12742
|
$\mathbf{L}^p$-boundedness of the Bochner-Riesz operator
|
In this paper, we give a new approach to the Bochner-Riesz summability. As a result, we show that the Bochner-Riesz operator $\mathbf{S}^δ, 0<\Reδ<{1\over 2}$ is bounded on $\mathbf{L}^p(\mathbb{R}^n)$ for ${n-1\over 2n}\leq {1\over p}\leq{n+1\over 2n}$.
| 2025-12-26
| 2025-12-29
|
[
"math.CA"
] |
Zipeng Wang
|
2512.22070
|
Next-to-leading order QCD corrections to electromagnetic production and decay of fully charm tetraquarks
|
We investigate the electromagnetic properties of the fully charm tetraquark states, particularly incorporating contributions from internal gluon radiations. The paper first presents analytical expressions for the next-to-leading-order (NLO) QCD corrections to the decay amplitudes of fully charm tetraquarks into two photons. It is found that the QCD corrections are significant for the $J^{PC}=0^{++}$ fully charm tetraquark decay process, whereas they are relatively small for the $J^{PC}=2^{++}$ fully charm tetraquark decay process. Subsequently, by considering photon-photon fusion in ultra-peripheral high-energy collisions of protons and nuclei and in electron-positron collision processes, we provide theoretical predictions for the production cross sections of fully-charm tetraquark states. The results presented in this work regarding the electromagnetic production and decay of fully charm tetraquarks shall be tested in current and future experiments.
| 2025-12-26
| 2025-12-29
|
[
"hep-ph",
"hep-ex",
"hep-lat"
] |
Xinran Liu, Yefan Wang, Ruilin Zhu
|
2512.22351
|
VULCAN: Tool-Augmented Multi Agents for Iterative 3D Object Arrangement
|
Despite the remarkable progress of Multimodal Large Language Models (MLLMs) in 2D vision-language tasks, their application to complex 3D scene manipulation remains underexplored. In this paper, we bridge this critical gap by tackling three key challenges in 3D object arrangement task using MLLMs. First, to address the weak visual grounding of MLLMs, which struggle to link programmatic edits with precise 3D outcomes, we introduce an MCP-based API. This shifts the interaction from brittle raw code manipulation to more robust, function-level updates. Second, we augment the MLLM's 3D scene understanding with a suite of specialized visual tools to analyze scene state, gather spatial information, and validate action outcomes. This perceptual feedback loop is critical for closing the gap between language-based updates and precise 3D-aware manipulation. Third, to manage the iterative, error-prone updates, we propose a collaborative multi-agent framework with designated roles for planning, execution, and verification. This decomposition allows the system to robustly handle multi-step instructions and recover from intermediate errors. We demonstrate the effectiveness of our approach on a diverse set of 25 complex object arrangement tasks, where it significantly outperforms existing baselines. Website: vulcan-3d.github.io
| 2025-12-26
| 2025-12-30
|
[
"cs.CV",
"cs.AI"
] |
Zhengfei Kuang, Rui Lin, Long Zhao, Gordon Wetzstein, Saining Xie, Sanghyun Woo
|
2512.21998
|
Multi-Satellite Multi-Stream Beamspace Massive MIMO Transmission
|
This paper studies multi-satellite multi-stream (MSMS) beamspace transmission, where multiple satellites cooperate to form a distributed multiple-input multiple-output (MIMO) system and jointly deliver multiple data streams to multi-antenna user terminals (UTs), and beamspace transmission combines earth-moving beamforming with beam-domain precoding. For the first time, we formulate the signal model for MSMS beamspace MIMO transmission. Under synchronization errors, multi-antenna UTs enable the distributed MIMO channel to exhibit higher rank, supporting multiple data streams. Beamspace MIMO retains conventional codebook based beamforming while providing the performance gains of precoding. Based on the signal model, we propose statistical channel state information (sCSI)-based optimization of satellite clustering, beam selection, and transmit precoding, using a sum-rate upper-bound approximation. With given satellite clustering and beam selection, we cast precoder design as an equivalent covariance decomposition-based weighted minimum mean square error (CDWMMSE) problem. To obtain tractable algorithms, we develop a closed-form covariance decomposition required by CDWMMSE and derive an iterative MSMS beam-domain precoder under sCSI. Following this, we further propose several heuristic closed-form precoders to avoid iterative cost. For satellite clustering, we enhance a competition-based algorithm by introducing a mechanism to regulate the number of satellites serving certain UT. Furthermore, we design a two-stage low-complexity beam selection algorithm focused on enhancing the effective channel power. Simulations under practical configurations validate the proposed methods across the number of data streams, receive antennas, serving satellites, and active beams, and show that beamspace transmission approaches conventional MIMO performance at lower complexity.
| 2025-12-26
| 2025-12-29
|
[
"eess.SP"
] |
Yafei Wang, Yiming Zhu, Vu Nguyen Ha, Wenjin Wang, Rui Ding, Symeon Chatzinotas, Björn Ottersten
|
2502.17260
|
Robust Federated Learning in Unreliable Wireless Networks: A Client Selection Approach
|
Federated learning (FL) has emerged as a promising distributed learning paradigm for training deep neural networks (DNNs) at the wireless edge, but its performance can be severely hindered by unreliable wireless transmission and inherent data heterogeneity among clients. Existing solutions primarily address these challenges by incorporating wireless resource optimization strategies, often focusing on uplink resource allocation across clients under the assumption of homogeneous client-server network standards. However, these approaches overlooked the fact that mobile clients may connect to the server via diverse network standards (e.g., 4G, 5G, Wi-Fi) with customized configurations, limiting the flexibility of server-side modifications and restricting applicability in real-world commercial networks. This paper presents a novel theoretical analysis about how transmission failures in unreliable networks distort the effective label distributions of local samples, causing deviations from the global data distribution and introducing convergence bias in FL. Our analysis reveals that a carefully designed client selection strategy can mitigate biases induced by network unreliability and data heterogeneity. Motivated by this insight, we propose FedCote, a client selection approach that optimizes client selection probabilities without relying on wireless resource scheduling. Experimental results demonstrate the robustness of FedCote in DNN-based classification tasks under unreliable networks with frequent transmission failures.
| 2025-12-26
| 2025-12-29
|
[
"cs.DC",
"cs.LG"
] |
Yanmeng Wang, Wenkai Ji, Jian Zhou, Fu Xiao, Tsung-Hui Chang
|
2512.22004
|
Solutions of 3D Reflection Equation from Quantum Cluster Algebra Associated with Symmetric Butterfly Quiver
|
We construct a new solution $(R,K)$ to the three-dimensional reflection equation, a boundary analogue of the tetrahedron equation. The $R$-operator is the one obtained by Sun, Terashima, Yagi, and the authors in 2024, involving four quantum dilogarithms with arguments in the $q$-Weyl algebra. The new $K$-operator similarly involves ten such quantum dilogarithms. Our approach is based on the quantum cluster algebra associated with the symmetric butterfly quiver on the wiring diagram of type C.
| 2025-12-26
| 2025-12-29
|
[
"math.QA",
"math-ph",
"math.MP",
"nlin.SI"
] |
Rei Inoue, Atsuo Kuniba
|
2512.14867
|
Link of the Zitterbewegung with the spin conductivity and the spin-textures of multiband systems
|
The Zitterbewegung phenomenon in multiband electronic systems is known to be subtly related to the charge conductivity, Berry curvature and the Chern number. Here we show that some spin-dependent properties as the optical spin conductivity, and intrinsic spin Hall conductivity are also entangled with the Zitterbewegung amplitudes. We also show that in multiband Dirac-type Hamiltonians, a direct link between the Zitterbewegung and the spin textures and spin transition amplitudes can be established. The later allow us to discern the presence or not of the Zitterbewegung oscillations by simply analyzing the spin or pseudo-spin textures. We provide examples of the applicability of our approach for Hamiltonian models that show the suppression of specific Zitterbewegung oscillations.
| 2025-12-26
| 2025-12-29
|
[
"cond-mat.mes-hall"
] |
F. Mireles, E. Ortiz
|
2210.11626
|
Optimal plug-in Gaussian processes for modeling derivatives
|
Derivatives are a key nonparametric functional in wide-ranging applications where the rate of change of an unknown function is of interest. In the Bayesian paradigm, Gaussian processes (GPs) are routinely used as a flexible prior for unknown functions, and are arguably one of the most popular tools in many areas. However, little is known about the optimal modeling strategy and theoretical properties when using GPs for derivatives. In this article, we study a plug-in strategy by differentiating the posterior distribution with GP priors for derivatives of any order. This practically appealing plug-in GP method has been previously perceived as suboptimal and degraded, but this is not necessarily the case. We provide posterior contraction rates for plug-in GPs and establish that they achieve optimal rates simultaneously for all derivative orders. We show that the posterior measure of the regression function and its derivatives, with the same choice of hyperparameter that does not depend on the order of derivatives, converges at the minimax optimal rate up to a logarithmic factor for functions in certain classes. We analyze a data-driven hyperparameter tuning method based on empirical Bayes, and show that it satisfies the optimal rate condition while maintaining computational efficiency. This article to the best of our knowledge provides the first positive result for plug-in GPs in the context of inferring derivative functionals, and leads to a practically simple nonparametric Bayesian method with optimal and adaptive hyperparameter tuning for simultaneously estimating the regression function and its derivatives. Simulations show competitive finite sample performance of the plug-in GP method. A climate change application for analyzing the global sea-level rise is discussed.
| 2025-12-26
| 2025-12-30
|
[
"math.ST",
"stat.TH"
] |
Zejian Liu, Meng Li
|
2512.21885
|
Spectral State Switching in Mrk 421: Results from the AstroSat LAXPC/SXT Observations
|
We carried a detailed time and flux resolved X-ray spectral analysis of the high-synchrotron-peaked blazar Mrk\,421 using simultaneous AstroSat and LAXPC20/SXT observations. The 100\,s binned LAXPC20 light curve obtained during 3--8 January 2017 reveals pronounced flux variability. The source exhibits a fractional variability amplitude of $F_{\mathrm{rms}} = 0.210 \pm 0.005$ in the SXT band and $F_{\mathrm{rms}} = 0.316 \pm 0.006$ in the LAXPC20 band. During this interval, the source reached a peak count rate of 122.94\,counts\,s$^{-1}$. This enabled us to carry flux resolved spectroscopy by selecting ten flux states, S1--S10 each having a width of 8\,counts\,s$^{-1}$. We noted that the spectra in these flux states are well described by a synchrotron-convolved broken power-law (BPL), which consistently provides a better fit than a log-parabola. The low-energy particle index (index before the break) is found to cluster around two discrete values across flux states indicating two spectra states in the source. The break energy consistently moves to high energy with increase in flux level in these states. Time-resolved spectroscopy (10-ks segments) confirms that the flux histogram is best modelled as a double lognormal distribution and the index histogram is bimodal. Inclusion of two additional long observations spanning 2017-2019 shows the same double-state behaviour on longer timescales. Together, the results indicate that Mrk\,421 routinely occupies two dominant spectral; in a leptonic synchrotron framework this can be explained by Gaussian-like fluctuations in acceleration conditions producing lognormal flux states.
| 2025-12-26
| 2025-12-29
|
[
"astro-ph.HE"
] |
Sikandar Akbar, Zahir Shah, Ranjeev Misra, Naseer Iqbal
|
2512.07471
|
First principle study of electronic, magnetic and thermoelectric properties of Co$_2$YPb (Y = Tc, Ti, Zr and Hf) full Heusler: Application to embedded automotive systems
|
In this study, theoretical investigation on structural, electronic, magnetic, elastic and thermoelectric properties of the full Heusler Co$_2$YPb (Y = Tc, Ti, Zr and Hf) alloys have been performed within density functional theory (DFT). The exchange and correlation potential is addressed using two approximations: the generalized gradient approximation (GGA) and the GGA augmented by the Tran--Blaha-modified Becke-Johnson (mBj-GGA) approximation, which provides a more accurate description of the energy band gap. The electronic and magnetic properties reveal that the full-Heusler alloys Co$_2$YPb (with Y = Tc, Ti, Zr, and Hf) display half-metallic ferromagnetic behavior. Furthermore, the elastic properties suggest that Co$_2$YPb are mechanically stable, with ductile characteristics. Full Heusler alloys P-type exhibit positive Seebeck coefficients and high ZT values, indicating good thermoelectric performance in terms of electrical and thermal conductivity. This leads us to the conclusions that these compounds are very interesting in improving the performance of embedded automotive systems and can also be used in spintronic devices.
| 2025-12-26
| 2025-12-29
|
[
"cond-mat.mtrl-sci"
] |
N. Saidi, A. Abbad, W. Benstaali, K. Bahnes
|
2512.22349
|
Human-like visual computing advances explainability and few-shot learning in deep neural networks for complex physiological data
|
Machine vision models, particularly deep neural networks, are increasingly applied to physiological signal interpretation, including electrocardiography (ECG), yet they typically require large training datasets and offer limited insight into the causal features underlying their predictions. This lack of data efficiency and interpretability constrains their clinical reliability and alignment with human reasoning. Here, we show that a perception-informed pseudo-colouring technique, previously demonstrated to enhance human ECG interpretation, can improve both explainability and few-shot learning in deep neural networks analysing complex physiological data.
We focus on acquired, drug-induced long QT syndrome (LQTS) as a challenging case study characterised by heterogeneous signal morphology, variable heart rate, and scarce positive cases associated with life-threatening arrhythmias such as torsades de pointes. This setting provides a stringent test of model generalisation under extreme data scarcity. By encoding clinically salient temporal features, such as QT-interval duration, into structured colour representations, models learn discriminative and interpretable features from as few as one or five training examples. Using prototypical networks and a ResNet-18 architecture, we evaluate one-shot and few-shot learning on ECG images derived from single cardiac cycles and full 10-second rhythms. Explainability analyses show that pseudo-colouring guides attention toward clinically meaningful ECG features while suppressing irrelevant signal components. Aggregating multiple cardiac cycles further improves performance, mirroring human perceptual averaging across heartbeats. Together, these findings demonstrate that human-like perceptual encoding can bridge data efficiency, explainability, and causal reasoning in medical machine intelligence.
| 2025-12-26
| 2025-12-30
|
[
"cs.CV",
"cs.AI",
"cs.HC",
"cs.LG"
] |
Alaa Alahmadi, Mohamed Hasan
|
2503.09626
|
Certainly Bot Or Not? Trustworthy Social Bot Detection via Robust Multi-Modal Neural Processes
|
Social bot detection is crucial for mitigating misinformation, online manipulation, and coordinated inauthentic behavior. While existing neural network-based detectors perform well on benchmarks, they struggle with generalization due to distribution shifts across datasets and frequently produce overconfident predictions for out-of-distribution accounts beyond the training data. To address this, we introduce a novel Uncertainty Estimation for Social Bot Detection (UESBD) framework, which quantifies the predictive uncertainty of detectors beyond mere classification. For this task, we propose Robust Multi-modal Neural Processes (RMNP), which aims to enhance the robustness of multi-modal neural processes to modality inconsistencies caused by social bot camouflage. RMNP first learns unimodal representations through modality-specific encoders. Then, unimodal attentive neural processes are employed to encode the Gaussian distribution of unimodal latent variables. Furthermore, to avoid social bots stealing human features to camouflage themselves thus causing certain modalities to provide conflictive information, we introduce an evidential gating network to explicitly model the reliability of modalities. The joint latent distribution is learned through the generalized product of experts, which takes the reliability of each modality into consideration during fusion. The final prediction is obtained through Monte Carlo sampling of the joint latent distribution followed by a decoder. Experiments on three real-world benchmarks show the effectiveness of RMNP in classification and uncertainty estimation, as well as its robustness to modality conflicts.
| 2025-12-26
| 2025-12-29
|
[
"cs.SI",
"cs.AI",
"cs.LG"
] |
Qi Wu, Yingguang Yang, hao liu, Hao Peng, Buyun He, Yutong Xia, Yong Liao
|
2512.21897
|
MMCTOP: A Multimodal Textualization and Mixture-of-Experts Framework for Clinical Trial Outcome Prediction
|
Addressing the challenge of multimodal data fusion in high-dimensional biomedical informatics, we propose MMCTOP, a MultiModal Clinical-Trial Outcome Prediction framework that integrates heterogeneous biomedical signals spanning (i) molecular structure representations, (ii) protocol metadata and long-form eligibility narratives, and (iii) disease ontologies. MMCTOP couples schema-guided textualization and input-fidelity validation with modality-aware representation learning, in which domain-specific encoders generate aligned embeddings that are fused by a transformer backbone augmented with a drug-disease-conditioned sparse Mixture-of-Experts (SMoE). This design explicitly supports specialization across therapeutic and design subspaces while maintaining scalable computation through top-k routing. MMCTOP achieves consistent improvements in precision, F1, and AUC over unimodal and multimodal baselines on benchmark datasets, and ablations show that schema-guided textualization and selective expert routing contribute materially to performance and stability. We additionally apply temperature scaling to obtain calibrated probabilities, ensuring reliable risk estimation for downstream decision support. Overall, MMCTOP advances multimodal trial modeling by combining controlled narrative normalization, context-conditioned expert fusion, and operational safeguards aimed at auditability and reproducibility in biomedical informatics.
| 2025-12-26
| 2025-12-29
|
[
"cs.LG",
"cs.AI"
] |
Carolina AparÃcio, Qi Shi, Bo Wen, Tesfaye Yadete, Qiwei Han
|
2512.21934
|
Double-Layered Silica-Engineered Fluorescent Nanodiamonds for Catalytic Generation and Quantum Sensing of Active Radicals
|
Fluorescent nanodiamonds (FNDs) hosting nitrogen-vacancy (NV) centers have attracted considerable attention for quantum sensing applications, particularly owing to notable advancements achieved in the field of weak magnetic signal detection in recent years. Here, we report a practical quantum-sensing platform for the controlled production and real-time monitoring of ultra-short-lived reactive free radicals using a double-layered silica modification strategy. An inner dense silica layer preserves the intrinsic properties of NV centers, while an outer porous silica layer facilitates efficient adsorption and stabilization of hydroxyl radicals and their precursor reactants. By doping this mesoporous shell with gadolinium (III) catalysts, we achieve sustained, light-free generation of hydroxyl radicals via catalytic water splitting, eliminating reliance on external precursors. The mechanism underlying this efficient radical generation is discussed in detail. The radical production is monitored in real time and in situ through spin-dependent T1 relaxometry of the NV centers, demonstrating stable and tunable radical fluxes, with concentration tunable across a continuous range from approximately 100 mM to molar levels by adjusting the catalyst condition. This study extends the technical application of nanodiamonds from relaxation sensing to the controlled synthesis of reactive free radicals, thereby providing robust experimental evidence to support the advancement of quantum sensing systems in intelligent manufacturing.
| 2025-12-26
| 2025-12-29
|
[
"quant-ph",
"physics.chem-ph"
] |
Jia Su, Zenghao Kong, Fei Kong, Xing Liu, Linyu Zeng, Zhecheng Wang, Zijian Zeng, Jie Liu, Jihu Su, Junhua Yuan, Guosheng Shi, Fazhan Shi
|
2503.17809
|
Poisson-Process Topic Model for Integrating Knowledge from Pre-trained Language Models
|
Topic modeling is traditionally applied to word counts without accounting for the context in which words appear. Recent advancements in large language models (LLMs) offer contextualized word embeddings, which capture deeper meaning and relationships between words. We aim to leverage such embeddings to improve topic modeling.
We use a pre-trained LLM to convert each document into a sequence of word embeddings. This sequence is then modeled as a Poisson point process, with its intensity measure expressed as a convex combination of $K$ base measures, each corresponding to a topic. To estimate these topics, we propose a flexible algorithm that integrates traditional topic modeling methods, enhanced by net-rounding applied before and kernel smoothing applied after. One advantage of this framework is that it treats the LLM as a black box, requiring no fine-tuning of its parameters. Another advantage is its ability to seamlessly integrate any traditional topic modeling approach as a plug-in module, without the need for modifications
Assuming each topic is a $β$-Hölder smooth intensity measure on the embedded space, we establish the rate of convergence of our method. We also provide a minimax lower bound and show that the rate of our method matches with the lower bound when $β\leq 1$. Additionally, we apply our method to several datasets, providing evidence that it offers an advantage over traditional topic modeling approaches.
| 2025-12-26
| 2025-12-30
|
[
"stat.ML",
"cs.LG",
"math.ST",
"stat.TH"
] |
Morgane Austern, Yuanchuan Guo, Zheng Tracy Ke, Tianle Liu
|
2512.14108
|
An integrable hierarchy associated with loop extension of $\mathbb{Z}_2^2$-graded $\mathfrak{osp}(1|2)$
|
A hierarchy of $\mathbb{Z}_2^2$-graded integrable equations is constructed using the loop extension of the $\mathbb{Z}_2^2$-graded Lie superalgebra $\mathfrak{osp}(1|2)$. This hierarchy includes $\mathbb{Z}_2^2$-graded extensions of the Liouville, sinh-Gordon, cosh-Gordon, and, in particular, the mKdV equations. The $\mathbb{Z}_2^2$-graded KdV equation is also derived from the $\mathbb{Z}_2^2$-mKdV equation via the Miura transformation. We present explicit formulas for the conserved charges of the $\mathbb{Z}_2^2$-KdV and $\mathbb{Z}_2^2$-mKdV equations. A distinctive feature of these $\mathbb{Z}_2^2$-graded integrable systems is the existence of conserved charges with nontrivial grading.
| 2025-12-26
| 2025-12-29
|
[
"math-ph",
"hep-th",
"math.MP",
"nlin.SI"
] |
N. Aizawa, I. Fujii, R. Ito
|
2511.04526
|
Generalizing Goodstein's theorem and Cichon's independence proof
|
We generalize Goodstein's theorem (Goodstein 1944) and Cichon's independence proof (Cichon 1983) to $Î ^1_1-\mathrm{CA}_0$ using results from (Wilken 2026). The method is generalizable to stronger notation systems that provide unique terms for ordinals and enjoy Bachmann property.
| 2025-12-26
| 2025-12-29
|
[
"math.LO"
] |
Gunnar Wilken
|
2512.22068
|
On the Ergodic Capacity for SIM-Aided Holographic MIMO Communications
|
We derive a novel closed-form lower bound on the ergodic capacity of holographic multiple-input multiple-output (HMIMO) systems enhanced by stacked intelligent metasurfaces (SIMs) under Rayleigh fading conditions. The proposed expression is valid for systems with a finite number of antennas and SIM elements and exhibits tightness throughout the whole signal-to-noise ratio (SNR) range. Furthermore, we conduct a comprehensive low-SNR analysis, offering meaningful observations on how key system parameters influence the capacity performance.
| 2025-12-26
| 2025-12-29
|
[
"cs.IT",
"math.IT"
] |
Anastasios Papazafeiropoulos, Ioannis Bartsiokas, Dimitra I. Kaklamani, Iakovos S. Venieris
|
2509.23343
|
Cosmological Prediction from the joint observation of MeerKAT and CSST at $z$ = 0.4 $\sim$ 1.2
|
Cross-correlating neutral hydrogen (HI) 21cm intensity mapping with galaxy surveys provides an effective probe of astrophysical and cosmological information. This work presents a cross-correlation analysis between MeerKAT single-dish HI intensity mapping and Chinese Space Station Survey Telescope (CSST) spectroscopic galaxy surveys in $z=0.4\sim1.2$, which will share a survey area of several thousand square degrees. Utilizing Jiutian-1G cosmological simulation, we simulate the observational data of MeerKAT and CSST with survey areas from $\sim1600$ to $600$ deg$^2$ at $z=0.5$, 0.7, and 1. The effects of beam pattern, polarization leakage, and different foregrounds in the MeerKAT HI intensity mapping are considered in the simulation. After employing foreground removal with the principal component analysis (PCA) method and performing signal compensation, we derive the cross-power spectra of MeerKAT and CSST. We perform the joint constraint using the CSST galaxy auto-power spectra and MeerKAT-CSST cross-power spectra with the least-squares fitting method. The constraint results show that, in the simulated survey area, the relative accuracy can achieve $6\%\sim 8\%$ for the parameter products $Ω_{\rm HI}b_{\rm HI}b_{g}r_{\mathrm{HI},g}$ and $Ω_{\rm HI}b_{\rm HI}r_{\mathrm{HI},g}$ at the three redshifts, which is $3\sim4$ times smaller than the current result. These findings indicate that the full MeerKAT-CSST joint observation with thousands of square degrees overlapping survey area can be a powerful probe of large-scale structure, and has the ability to provide information on cosmic evolution of HI and galaxies in a wide redshift range.
| 2025-12-26
| 2025-12-29
|
[
"astro-ph.CO"
] |
Yu-Er Jiang, Yan Gong, Qi Xiong, Wenxiang Pei, Yun Liu, Furen Deng, Zi-yan Yuwen, Meng Zhang, Xingchen Zhou, Xuelei Chen, Yin-Zhe Ma, Qi Guo, Bin Yue
|
2405.13741
|
Existence in NSOP$_1$ theories
|
We show that Kim-forking satisfies existence in all NSOP$_1$ theories.
| 2025-12-26
| 2025-12-30
|
[
"math.LO"
] |
Byunghan Kim, Joonhee Kim, Hyoyoon Lee
|
2511.00977
|
Modeling Microenvironment Trajectories on Spatial Transcriptomics with NicheFlow
|
Understanding the evolution of cellular microenvironments in spatiotemporal data is essential for deciphering tissue development and disease progression. While experimental techniques like spatial transcriptomics now enable high-resolution mapping of tissue organization across space and time, current methods that model cellular evolution operate at the single-cell level, overlooking the coordinated development of cellular states in a tissue. We introduce NicheFlow, a flow-based generative model that infers the temporal trajectory of cellular microenvironments across sequential spatial slides. By representing local cell neighborhoods as point clouds, NicheFlow jointly models the evolution of cell states and spatial coordinates using optimal transport and Variational Flow Matching. Our approach successfully recovers both global spatial architecture and local microenvironment composition across diverse spatiotemporal datasets, from embryonic to brain development.
| 2025-12-26
| 2025-12-29
|
[
"cs.LG",
"q-bio.QM"
] |
Kristiyan Sakalyan, Alessandro Palma, Filippo Guerranti, Fabian J. Theis, Stephan Günnemann
|
2512.22321
|
Interface Modeling of Perovskite Polymer Heterostructures for Enhanced Charge Transfer Efficiency in Hybrid Photovoltaic Materials
|
Perovskite solar cells (PSCs) based on methylammonium lead iodide (MAPbI3) exhibit remarkable photovoltaic performance, where interface engineering with hole transport layers (HTLs) is crucial for optimizing charge transfer and device efficiency. In this work, we present a density functional theory (DFT) study of the MAPbI3/poly(3-hexylthiophene) (P3HT) hybrid interface, focusing on the role of perovskite surface termination in determining interfacial stability and electronic structure. We model MAI- and PbI-terminated MAPbI3 surfaces interfaced with P3HT and compare their interfacial electronic properties. Electronic structure calculations reveal distinct differences in orbital hybridization and band alignment: the MAI/m-P3HT interface exhibits weak coupling, whereas the PbI/m-P3HT interface shows stronger hybridization and enhanced charge transfer. Band alignment confirms type-II, hole-selective character in both cases, with more pronounced valence band maximum adjustment for PbI. Charge difference maps, Bader analysis, and local density of states consistently indicate higher charge transfer and stronger electronic coupling for PbI termination. Electrostatic potential offsets and transport parameters further highlight termination-dependent differences, with lighter effective masses at PbI/m-P3HT and higher hole velocity at MAI/m-P3HT. These findings provide theoretical insight into interfacial charge transport mechanisms and offer guidelines for optimizing perovskite-organic hybrid solar cells.
| 2025-12-26
| 2025-12-30
|
[
"cond-mat.mtrl-sci",
"cond-mat.mes-hall"
] |
Somayyeh Alidoust, V. Ongun Ãzçelik
|
2512.21899
|
The First X-ray Polarimetry of GRS 1739--278 Reveals Its Rapidly Spinning Black Hole
|
We present a joint spectro-polarimetric analysis of the black hole X-ray binary GRS~1739--278 during its 2025 mini-outburst, using simultaneous observations from \ixpe\ and \nustar. The \ixpe\ data show a polarization degree of ${\rm PD} = (2.3 \pm 0.4)\%$ and a polarization angle of ${\rm PA} = 62^\circ \pm 5^\circ$ in the 2--8~keV range. The model-independent analysis reveals that the PD increases from $\sim 2\%$ at 2~keV to $\sim 10\%$ in the 6--8~keV band, while the PA remains stable across the \ixpe\ band within statistical uncertainties. Broadband spectral modeling of the combined \ixpe\ and \nustar\ datasets shows that hard Comptonization contributes negligibly in this soft-state observation, while a substantial reflected component is required in addition to the thermal disk emission. We then model the \ixpe\ Stokes spectra using the \texttt{kynbbrr} model. The best-fitting results indicate that high-spin configurations enhance the contribution of reflected returning radiation, which dominates the observed polarization properties. From the \texttt{kynbbrr} modeling, we infer an extreme black hole spin of $a = 0.994^{+0.004}_{-0.003}$ and a system inclination of $i = 54^\circ{}^{+8^\circ}_{-4^\circ}$. Owing to the large contribution from returning radiation, the observed polarization direction is nearly parallel to the projected system axis, the position angle of which is predicted to be $58^\circ \pm 4^\circ$. Our results demonstrate that X-ray polarimetry, combined with broadband spectroscopy, directly probes the geometry and relativistic effects in accretion disks around stellar-mass black holes.
| 2025-12-26
| 2025-12-29
|
[
"astro-ph.HE"
] |
Qing-Chang Zhao, Michal Dovciak, Han-Cheng Li, Lian Tao, Hua Feng, Federico Vincentelli, Giorgio Matt, Philip Kaaret, Shuang-Nan Zhang
|
2509.09386
|
Feedback-Controlled Beam Pattern Measurement Method Using a Power-Variable Calibration Source for Cosmic Microwave Background Telescopes
|
We demonstrate a novel beam pattern measurement method for the side lobe characterization of cosmic microwave background telescopes. The method employs a power-variable artificial microwave source under feedback control from the detector under test on the telescope. It enables us to extend the dynamic range of the beam pattern measurement without introducing nonlinearity effects from the detector. We conducted a laboratory-based proof-of-concept experiment, measuring the H-plane beam pattern of a horn antenna coupled to a diode detector at 81 GHz. We gained an additional dynamic range of 60.3 dB attributed to the feedback control. In addition, we verified the measurement by comparing it with other reference measurements obtained using conventional methods. The method is also applicable to general optical measurements requiring a high dynamic range to detect subtle nonidealities in the characteristics of optical devices.
| 2025-12-26
| 2025-12-29
|
[
"astro-ph.IM"
] |
Haruaki Hirose, Masaya Hasegawa, Daisuke Kaneko, Taketo Nagasaki, Ryota Takaku, Tijmen de Haan, Satoru Takakura, Takuro Fujino
|
2512.18475
|
Research on a hybrid LSTM-CNN-Attention model for text-based web content classification
|
This study presents a hybrid deep learning architecture that integrates LSTM, CNN, and an Attention mechanism to enhance the classification of web content based on text. Pretrained GloVe embeddings are used to represent words as dense vectors that preserve semantic similarity. The CNN layer extracts local n-gram patterns and lexical features, while the LSTM layer models long-range dependencies and sequential structure. The integrated Attention mechanism enables the model to focus selectively on the most informative parts of the input sequence. A 5-fold cross-validation setup was used to assess the robustness and generalizability of the proposed solution. Experimental results show that the hybrid LSTM-CNN-Attention model achieved outstanding performance, with an accuracy of 0.98, precision of 0.94, recall of 0.92, and F1-score of 0.93. These results surpass the performance of baseline models based solely on CNNs, LSTMs, or transformer-based classifiers such as BERT. The combination of neural network components enabled the model to effectively capture both fine-grained text structures and broader semantic context. Furthermore, the use of GloVe embeddings provided an efficient and effective representation of textual data, making the model suitable for integration into systems with real-time or near-real-time requirements. The proposed hybrid architecture demonstrates high effectiveness in text-based web content classification, particularly in tasks requiring both syntactic feature extraction and semantic interpretation. By combining presented mechanisms, the model addresses the limitations of individual architectures and achieves improved generalization. These findings support the broader use of hybrid deep learning approaches in NLP applications, especially where complex, unstructured textual data must be processed and classified with high reliability.
| 2025-12-26
| 2025-12-29
|
[
"cs.CL",
"cs.LG"
] |
Mykola Kuz, Ihor Lazarovych, Mykola Kozlenko, Mykola Pikuliak, Andrii Kvasniuk
|
2505.01433
|
Enhancing TCR-Peptide Interaction Prediction with Pretrained Language Models and Molecular Representations
|
Understanding the binding specificity between T-cell receptors (TCRs) and peptide-major histocompatibility complexes (pMHCs) is central to immunotherapy and vaccine development. However, current predictive models struggle with generalization, especially in data-scarce settings and when faced with novel epitopes. We present LANTERN (Large lAnguage model-powered TCR-Enhanced Recognition Network), a deep learning framework that combines large-scale protein language models with chemical representations of peptides. By encoding TCR \b{eta}-chain sequences using ESM-1b and transforming peptide sequences into SMILES strings processed by MolFormer, LANTERN captures rich biological and chemical features critical for TCR-peptide recognition. Through extensive benchmarking against existing models such as ChemBERTa, TITAN, and NetTCR, LANTERN demonstrates superior performance, particularly in zero-shot and few-shot learning scenarios. Our model also benefits from a robust negative sampling strategy and shows significant clustering improvements via embedding analysis. These results highlight the potential of LANTERN to advance TCR-pMHC binding prediction and support the development of personalized immunotherapies.
| 2025-12-26
| 2025-12-29
|
[
"q-bio.QM",
"cs.CL",
"cs.LG"
] |
Cong Qi, Hanzhang Fang, Siqi jiang, Tianxing Hu, Zhi Wei
|
2512.21977
|
Repeat times and a two-weight UST model
|
We study a model of random weighted uniform spanning trees on the complete graph with $n$ vertices, where each edge is assigned a weight of $n^{1+γ}$ with probability $1/n$ and $1$ otherwise. Whenever $γ$ is large enough, we prove that the diameter of the resulting tree is typically of order $n^{1/3} \log n$, up to a $\log \log n$ correction. Our approach uses estimates on repeat times for selecting components in a critical ErdÅs-Rényi graph, as well as concentration bounds on the sums of diameters of these components.
| 2025-12-26
| 2025-12-29
|
[
"math.PR",
"math.CO"
] |
Umberto De Ambroggio, Luca Makowiec
|
2512.21910
|
Fano Fibrations and Twisted Kähler-Einstein Metrics II: The Kähler-Ricci Flow
|
This is the second of two papers studying both the geometric structure of Fano fibrations and the application to Kähler-Ricci flows developing a singularity in finite time. We assume that the Kähler-Ricci flow on a compact Kähler manifold has a rational initial metric and develops a singularity in finite time such that the manifold admits a Fano fibration structure. Moreover, it is assumed that the volume form of the flow collapses uniformly at the rate of $C^{-1}(T-t)^{n-m} Ω\leq Ï(t)^n\leq C(T-t)^{n-m}Ω$. Under this setting, a diameter bound is obtained in any compact set away from singular fibres and the diameter of the fibres is proven to collapse at the optimal rate $\sqrt{T-t}$. Furthermore, several precise $C^0$-estimates are proven for the potential of the complex Monge-Ampere flow which involve the potentials of singular twisted Kähler-Einstein metrics on the base variety from part I. Finally, in the case of Kähler-Einstein Fano fibres, we deduce Type I scalar curvature in any compact set away from singular fibres and globally for a submersion.
| 2025-12-26
| 2025-12-29
|
[
"math.DG",
"math.CV"
] |
Alexander Bednarek
|
2511.16953
|
Merging RLBWTs adaptively
|
We show how we can merge two run-length compressed Burrows-Wheeler Transforms (RLBWTs) into a run-length compressed extended Burrows-Wheeler Transform (eBWT) in $O (r)$ space and $O ((r + L) \log (m + n))$ time, where $m$ and $n$ are the lengths of the uncompressed strings, $r$ is the number of runs in the final eBWT and $L$ is the sum of the longest common prefix (LCP) values at the beginnings of those runs.
| 2025-12-26
| 2025-12-29
|
[
"cs.DS"
] |
Travis Gagie
|
2512.22350
|
Magneto-Optical Trapping of a Metal Hydride Molecule
|
We demonstrate a three-dimensional magneto-optical trap (MOT) of a metal hydride molecule, CaH. We are able to scatter $\sim$$10^{4}$ photons with vibrational loss covered up to vibrational quantum number $ν=2$. This allows us to laser slow the molecular beam near zero velocity with a "white-light" technique and subsequently load it into a radio-frequency MOT. The MOT contains 230(40) molecules, limited by beam source characteristics and predissociative loss of CaH. The temperature of the MOT is below one millikelvin. The predissociative loss mechanism could, in turn, facilitate controlled dissociation of the molecule, offering a possible route to optical trapping of hydrogen atoms for precision spectroscopy.
| 2025-12-26
| 2025-12-30
|
[
"physics.atom-ph",
"cond-mat.quant-gas",
"quant-ph"
] |
Jinyu Dai, Benjamin Riley, Qi Sun, Debayan Mitra, Tanya Zelevinsky
|
2512.07849
|
AI Urban Scientist: Multi-Agent Collaborative Automation for Urban Research
|
Urban research aims to understand how cities operate and evolve as complex adaptive systems. With the rapid growth of urban data and analytical methodologies, the central challenge of the field has shifted from data availability to the integration of heterogeneous data into coherent, verifiable urban knowledge through multidisciplinary approaches. Recent advances in AI, particularly the emergence of large language models (LLMs), have enabled the development of AI scientists capable of autonomous reasoning, hypothesis generation, and data-driven experimentation, demonstrating substantial potential for autonomous urban research. However, most general-purpose AI systems remain misaligned with the domain-specific knowledge, methodological conventions, and inferential standards required in urban studies. Here, we introduce the AI Urban Scientist, a knowledge-driven multi-agent framework designed to support autonomous urban research. Grounded in hypotheses, peer-review feedback, datasets, and research methodologies distilled from large-scale prior studies, the system constructs structured domain knowledge that guides LLM-based agents to automatically generate hypotheses, identify and integrate multi-source urban datasets, conduct empirical analyses and simulations, and iteratively refine analytical methods. Through this process, the framework synthesizes new insights in urban science and accelerates the urban research lifecycle.
| 2025-12-26
| 2025-12-29
|
[
"cs.CY",
"cs.CL",
"cs.MA"
] |
Tong Xia, Jiankun Zhang, Ruiwen You, Ao Xu, Linghao Zhang, Tengyao Tu, Jingzhi Wang, Jinghua Piao, Yunke Zhang, Fengli Xu, Yong Li
|
2512.21818
|
Analyzing Code Injection Attacks on LLM-based Multi-Agent Systems in Software Development
|
Agentic AI and Multi-Agent Systems are poised to dominate industry and society imminently. Powered by goal-driven autonomy, they represent a powerful form of generative AI, marking a transition from reactive content generation into proactive multitasking capabilities. As an exemplar, we propose an architecture of a multi-agent system for the implementation phase of the software engineering process. We also present a comprehensive threat model for the proposed system. We demonstrate that while such systems can generate code quite accurately, they are vulnerable to attacks, including code injection. Due to their autonomous design and lack of humans in the loop, these systems cannot identify and respond to attacks by themselves. This paper analyzes the vulnerability of multi-agent systems and concludes that the coder-reviewer-tester architecture is more resilient than both the coder and coder-tester architectures, but is less efficient at writing code. We find that by adding a security analysis agent, we mitigate the loss in efficiency while achieving even better resiliency. We conclude by demonstrating that the security analysis agent is vulnerable to advanced code injection attacks, showing that embedding poisonous few-shot examples in the injected code can increase the attack success rate from 0% to 71.95%.
| 2025-12-26
| 2025-12-29
|
[
"cs.SE",
"cs.MA"
] |
Brian Bowers, Smita Khapre, Jugal Kalita
|
2512.19415
|
Non-Contrast CT Esophageal Varices Grading through Clinical Prior-Enhanced Multi-Organ Analysis
|
Esophageal varices (EV) represent a critical complication of portal hypertension, affecting approximately 60% of cirrhosis patients with a significant bleeding risk of ~30%. While traditionally diagnosed through invasive endoscopy, non-contrast computed tomography (NCCT) presents a potential non-invasive alternative that has yet to be fully utilized in clinical practice. We present Multi-Organ-COhesion Network++ (MOON++), a novel multimodal framework that enhances EV assessment through comprehensive analysis of NCCT scans. Inspired by clinical evidence correlating organ volumetric relationships with liver disease severity, MOON++ synthesizes imaging characteristics of the esophagus, liver, and spleen through multimodal learning. We evaluated our approach using 1,631 patients, those with endoscopically confirmed EV were classified into four severity grades. Validation in 239 patient cases and independent testing in 289 cases demonstrate superior performance compared to conventional single organ methods, achieving an AUC of 0.894 versus 0.803 for the severe grade EV classification (G3 versus <G3) and 0.921 versus 0.793 for the differentiation of moderate to severe grades (>=G2 versus <G2). We conducted a reader study involving experienced radiologists to further validate the performance of MOON++. To our knowledge, MOON++ represents the first comprehensive multi-organ NCCT analysis framework incorporating clinical knowledge priors for EV assessment, potentially offering a promising non-invasive diagnostic alternative.
| 2025-12-26
| 2025-12-29
|
[
"cs.CV"
] |
Xiaoming Zhang, Chunli Li, Jiacheng Hao, Yuan Gao, Danyang Tu, Jianyi Qiao, Xiaoli Yin, Le Lu, Ling Zhang, Ke Yan, Yang Hou, Yu Shi
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.