arXiv subjects

Ke Chen

Publications and source records attributed to Ke Chen.

At least 37 records · Page 2Linked to original sources

arXiv2026

Holographic dark energy from generalized entropies and extended infrared cut-offs

We study holographic dark energy models constructed from generalized entropy functionals and local infrared cut-offs. Three entropies (Barrow and the three and five-parameter generalized entropies) are each combined with the Granda-Oliveros (GO) cut-off and two extensions, one adding a $γH^{-2}$ term and one a $γ\ddot{H}/H$ term, the latter raising the dark-energy evolution equation to second order. The nine resulting models are constrained against DESI DR2 BAO, Pantheon+ and a Planck compressed CMB likelihood through an MCMC analysis, and compared using the Akaike and Bayesian information criteria. The Hubble constant is stable across all models. The $H^{-2}$ term is unconstrained in every case, as it becomes relevant only in the far future, and never improves on the GO cut-off. The $\ddot{H}/H$ term, instead, is tightly constrained for Barrow and $S_5$, where it improves the fit substantially, but not for $S_3$. The best fit is obtained for Barrow entropy with the $\ddot{H}/H$ cut-off, yet the Barrow exponent remains compatible with zero in all cases. The data thus favour a higher-order infrared cut-off, while showing no significant evidence for a departure from the Bekenstein-Hawking area law. We also investigate the dark energy equation of state (EoS) of $S_3$ and $S_5$, discovering quintessence-like behavior and phantom crossing.

gr-qc↗

arXiv2026

QCATS: Query Context-Aware Transformer Slicing for Efficient Predictive Query Processing

In-database predictive query processing increasingly applies Transformer-based models within relational pipelines. However, existing in-database inference typically exposes only tuple-level model inputs to the inference runtime, leaving relational predicates and metadata statistics invisible to neural execution planning. In this paper, we propose QCATS, a query context-aware transformer slicing framework that enables efficient sparse inference inside database systems. QCATS executes at query granularity: instead of routing individual tokens or tuples during inference, it uses query predicates and metadata statistics to pre-select context-aligned FFN slices before model execution. The framework comprises offline expert construction and lightweight query-level routing that dynamically selects experts during execution. QCATS further introduces system optimizations, including asynchronous CPU-GPU pipelines and routing-aware batching. Experiments on four predictive-query workloads with BERT-base and Qwen-0.6B show that QCATS achieves up to 4.42x latency reduction while preserving prediction accuracy comparable to dense baselines.

cs.DB↗

arXiv2026

TAC: Timestamped Audio Captioning

Large Audio Language Models struggle to disentangle overlapping events in complex acoustic scenes, yielding temporally inconsistent captions and frequent hallucinations. We introduce Timestamped Audio Captioner (TAC), a model that produces temporally grounded audio descriptions at varying degrees of detail and resolution. TAC is trained with a synthetic data pipeline that constructs challenging and dynamic mixtures from real-world audio sources, enabling robust learning under realistic polyphonic conditions. Across event detection and dense captioning, TAC outperforms all competing methods, with a low hallucination rate and accurate temporal grounding. We also introduce TAC-V, an audio-visual pipeline to generate semantically rich audio-visual descriptions. We then show that TAC and TAC-V serves as a "semantic bridge" for a text-only reasoner: a simple TAC$\rightarrow$LLM and TAC-V$\rightarrow$LLM cascade achieves state-of-the-art scores on benchmarks for both audio (MMAU-Pro, MMSU, MMAR) and audio-visual (DailyOmni, VideoHolmes) understanding and reasoning respectively.

cs.SD↗

arXiv2026

Can Agents Trust Their Skills? Uncovering Unsafe Chains of Trust in Skill-Based LLM Agents

LLM agents increasingly rely on installable skills, which are packages of instructions, code, and resources that equip them with task-specific capabilities and, once installed, can be automatically invoked across subsequent user tasks. This creates a chain of trust in which users delegate authority to agents, while agent frameworks admit skill-provided content into the agents' context with insufficient validation, allowing malicious skills to influence agent behavior under that delegated authority. Yet, little is known about whether this trust model adequately constrains untrusted skill content before it reaches security-sensitive operations, or how frequently such trust violations arise in real-world agents. We present TrustProbe, a framework for uncovering unsafe chains of trust in skill-based LLM agents. First, TrustProbe analyzes agent source code to identify source-to-sink call paths from skill-controlled inputs to security-sensitive operations. Second, it generates semantically realistic SKILL.md seeds with injected canaries and evolves them through feedback-guided scheduling and mutation. Finally, it validates vulnerabilities using an oracle that confirms attacker-controlled flows and verifies observable harm. Across 11 open-source agents, eight with more than 10,000 GitHub stars, TrustProbe identifies 104 taint-style vulnerabilities. Validation on a large corpus of real-world skills collected from public hubs such as ClawHub further shows that 25.1% of skill-agent trials exercise the identified vulnerable paths, with payload injection successfully weaponizing 15 of the vulnerabilities. These results reveal a systematic trust failure in skill-based LLM agents: untrusted skill content can reach security-sensitive operations and exercise authority delegated by users to their agents.

cs.CR↗

arXiv2026

A visual large language foundational model for medical image recognition using clinician-contributed online resources

Large language models (LLMs) have demonstrated strong capabilities across diverse domains, showing considerable potential in medicine. However, their application in medical settings remains limited by the scarcity of visual question answering (VQA) datasets that capture clinical reasoning and explicit image-text alignment. Here, we leverage de-identified medical images and expert commentaries shared through clinician-oriented online resources. By combining an advanced LLM with clinician-in-the-loop verification, we established a rigorous pipeline to construct ThoughtMed-1M, a long-form medical VQA dataset containing over one million VQA pairs and designed to capture structured clinical reasoning and medical image-text alignment. To demonstrate its utility, we developed a FOundational LLM Trained on ThoughtMed-1M (FOLTMed). FOLTMed achieved state-of-the-art performance across 42 medical VQA benchmark datasets, with a macro accuracy of 85.4 percent. It also generated more clinically coherent responses on the ThoughtMed-1M test set, outperforming state-of-the-art models by 3 to 5 percent across factuality and similarity metrics, highlighting a scalable paradigm for advancing research on clinically grounded multimodal LLMs.

cs.AI↗

arXiv2026

Evaluation Is All You Need for Multi-Modal Autonomous Driving

Multi-modal planning is promising for autonomous driving by representing multiple plausible behaviors in ambiguous and long-tail scenarios. Existing methods mainly focus on improving trajectory multi-modality, enhancing trajectory representations, or reshaping the candidate distribution. Nevertheless, we identify a pronounced generation-evaluation asymmetry in multi-modal planning: despite strong oracle performance, existing planners often fail to reliably select the best available candidate, leaving substantial planning potential unrealized. To address this challenge, we propose iDriveVLA, a multi-modal planning framework that improves the candidate trajectory space while enabling more reliable and context-aware trajectory evaluation. Specifically, iDriveVLA introduces a unified trajectory evaluator comprising a Safety-aware Scorer for quality and risk estimation, together with a VLM-guided Modulator for scene-adaptive criterion weighting. We further develop an oracle-aligned progressive training strategy consisting of candidate imitation pretraining, candidate space refinement, and semantic ranking alignment. On the public NAVSIM v1 leaderboard, iDriveVLA achieves a new state-of-the-art performance of 94.95 PDMS, surpassing the human-expert reference.

cs.RO↗

arXiv2026

Janus Dipoles: Fundamentals, Realizations, and Emerging Applications

The Janus dipole - featuring orthogonally oriented electric and magnetic dipoles with a 90-degree phase difference - has emerged as a powerful paradigm for wave manipulation. Unlike traditional Huygens dipoles used for directional control, this unique configuration exhibits strongly asymmetric, face-selective near-field behavior while maintaining a quasi-isotropic far-field radiation pattern. These remarkable properties make the Janus dipole an essential platform for directional wave shaping, with wide-ranging applications in on-chip photonics, quantum interactions, and wireless power transfer. This review systematically traces the rapid development of the Janus dipole from its foundational theoretical inception to its diverse implementation platforms across optical, microwave, and acoustic frequencies. In this paper, we explore the governing principles, classify realization strategies into passive Janus dipoles, active Janus dipoles, and advanced near-field coupling control, and highlight emerging frontiers. By bridging foundational electrodynamics with advanced device engineering, this paper serves as an essential reference and roadmap for researchers designing next-generation, highly integrated, and compact wave-manipulation systems.

physics.app-ph↗

arXiv2026

ParA-LLM: A Unified Approach to Paralinguistic and Acoustic Speech Understanding

Recent advances in Audio LLMs have achieved human-level speech recognition, yet existing systems struggle to capture paralinguistic aspects such as speaker traits, expressive variations, and environmental acoustic conditions. To address this, we design a framework of 22 paralinguistic characteristics and create a dataset of over 1.2M Audio-QA pairs. We develop ParA-LLM, trained with a two-stage curriculum: first on single-attribute questions to build foundational knowledge, then on multi-attribute questions for joint reasoning over speaker and acoustic characteristics. We also release ParA-Bench, a benchmark of 6,000 multiple-choice questions across speaker-speech, acoustic, and mixed categories, where frontier models like GPT-4o-Audio achieve only 36% accuracy. ParA-LLM surpasses state-of-the-art Audio LLMs like GPT-4o-Audio by 7.5% on ParA-Bench, with additional gains of 1.13% on MMAU-Pro Speech and 7.49% on MMAR Speech.

cs.SD↗

arXiv2026

Discontinuity of the Vlasov--Poisson Flow in $L_x^pL_v^\infty$

We prove that solution maps of the Vlasov--Poisson equation are discontinuous at the zero initial datum in $X_p=L_x^pL_v^\infty$ for every $1\leq p<\infty$, in $\mathbb{R}^{d} \times \mathbb{R}^{d}$ with $d \le 3$, and for both attractive and repulsive interactions. For every $T>0$, the trajectory of the solution in $L^\infty([0,T];X_p)$ is discontinuous at zero, and for every sufficiently small fixed $t>0$, the fixed-time map with values in $X_p$ is also discontinuous at zero. The counterexamples are smooth, nonnegative, bounded by one, and supported in a common compact subset of phase space. Their mass and initial $X_p$ norm tend to zero, whereas their solution norm at the observation time is at least one. The construction places narrow spatial packets on a lattice. Free transport allows a different velocity to select a packet at each spatial point, while a smallness estimate on the field ensures that the nonlinear characteristics are close to that of the free transport. For the same families, the phase-space $L^q$ norms converge to zero uniformly in time for every $1\leq q<\infty$.

math.AP↗

arXiv2026

Interfacial Engineering Enabled High-Resolution Stretchable Metal-Level Conductive Lines and Transparent Conductor

As stretchable electronics advance toward higher integration density and finer feature sizes, stretchable conductors, which serve as the architectural backbone of these electronics, must be scaled down accordingly. However, achieving both high stretchability and high electrical conductivity at high resolution remains a challenge using existing conductive materials. Here, through material design and interfacial engineering, we develop a thiol-functionalized conducting polymer/gold hybrid stack that can be patterned down to 4 micrometer linewidths using standard photolithography, while maintaining high stretchability, metal-like conductivity (>40,000 S/cm), and environmental stability. Leveraging this capability, we demonstrate grid-based stretchable transparent electrodes that surpass the figure-of-merit of indium tin oxide, as well as a 1000-pixel-per-inch image interconnected with stretchable lines. This work helps to broaden the scope of next-generation functional soft electronics, including e-skins and bioelectronics.

cond-mat.mtrl-sci↗

arXiv2026

Failure of analyticity-radius growth in energy-canceling fluid models

We prove that exact quadratic energy cancellation alone does not force growth of the spatial analyticity radius. On $\mathbb{T}^3$, we construct an explicit symmetric, translation-invariant, first-order bilinear operator $Q$ that preserves the divergence-free class and, for every smooth real-valued divergence-free vector field $v$, satisfies $ \int_{\mathbb{T}^3} Q(v,v)\cdot v\,dx=0.$ For every prescribed sufficiently small time $T>0$, the equation $\partial_t u-Δu=Q(u,u)$ admits a global smooth, real-valued, mean-zero, divergence-free solution $u$ such that $\operatorname{rad}(u(0))=\operatorname{rad}(u(T))=1$. The construction reduces the dynamics on an invariant cyclic-shear class to viscous Burgers and tunes a Cole-Hopf heat profile so that its nearest complex zero returns to its initial distance from the real torus. For every $1<α<2$ and every prescribed sufficiently small $T>0$, we also construct a symmetric sparse frequency set, its associated Fourier projection, and trigonometric-polynomial initial data for the projected dissipative surface quasi-geostrophic equation. The resulting unique global smooth solution $θ$ has infinite analyticity radius initially but satisfies $0<\operatorname{rad}(θ(T))\leq1$. An additively separated Fourier cascade yields coefficientwise exponential lower bounds, while uniform comparison estimates control the feedback interactions. Thus entire analyticity need not persist even from trigonometric-polynomial data. Together, the two constructions show that an exact energy identity alone does not determine the frequency geometry governing analyticity-radius growth.

math.AP↗

arXiv2026

OccAnyScene: Towards Unified Indoor-Outdoor 3D Occupancy Prediction

3D occupancy prediction is fundamental to scene understanding, yet existing 3D semantic occupancy methods are typically specialized to fixed scene types and occupancy protocols. We introduce Cross-Scene 3D Semantic Occupancy Prediction, a new task setting which requires a single model to handle heterogeneous indoor and outdoor scenes with varying cameras, spatial ranges, voxel specifications, and semantic taxonomies. This setting poses a fundamental challenge: achieving metric-consistent yet scene-adaptive image-to-3D lifting across varying camera configurations and scene scales. To address this challenge, we propose OccAnyScene, a pixel-frustum-centered Gaussian framework built upon a pretrained depth foundation model. Specifically, the framework employs Pixel-Aligned Frustum Feature Aggregation to construct a camera-aware frustum query for each feature pixel, and Frustum-Parameterized Gaussian Construction to decode each query into multiple Gaussians whose positions and sizes are constrained by the predicted pixel depth and corresponding frustum geometry. OccAnyScene sets new state-of-the-art results, achieving 59.92% mIoU on the indoor Occ-ScanNet and 23.06% mIoU on the outdoor SurroundOcc-nuScenes.

cs.CV↗

arXiv2026

CoLT-Drive: Counterfactual Long-Tail Benchmarking and Knowledge-Preserving Adaptation for Driving Affordance Prediction

Long-tail autonomous driving failures are often framed as rare-object recognition errors. We argue that this view is incomplete: the decision-critical question is not only whether a model recognizes an unusual object, but whether it infers how that object changes the ego vehicle's feasible high-level actions. We formalize this problem as decision-level driving affordance prediction, where a model maps a front-view image, ego-motion history, and navigation command to a structured longitudinal--lateral meta-action. To evaluate this capability, we introduce CoLT-Drive, a 3,536-sample counterfactual long-tail benchmark that inserts rare objects into otherwise fixed driving scenes and measures whether models predict acceptable action pairs. To improve deployable small VLMs, we propose KPA, a knowledge-preserving adaptation framework that combines structured perception-to-decision prompting, SLERP-based expert merging, and RegMoE, a regime-aware LoRA mixture-of-experts module. KPA preserves the pretrained model's open-world knowledge while allocating lightweight adaptation capacity to different driving decision regimes. Experiments on an in-domain driving split and CoLT-Drive show that KPA achieves 60.8\% pair accuracy on CoLT-Drive, outperforming the pretrained Qwen3-VL-2B baseline (50.3\%) and LoRA SFT (32.4\%) while maintaining competitive in-domain accuracy. Our benchmark and code are available at https://huggingface.co/datasets/tangzx2024/CoLT-Drive and https://github.com/tangzhengxu/CoLT-Drive.

cs.CV↗

arXiv2026

Beyond Textual Chain-of-Thought: A Survey on Action-Grounded Reasoning in Autonomous Driving

Chain-of-thought (CoT) reasoning powers generative models by eliciting intermediate steps before producing an answer. In autonomous driving, the answer is a continuous action. Thus its reasoning must share the same spatiotemporal structure as the physical world. This survey studies the resulting shift from textual CoT to action-grounded reasoning. Surveying 171 papers, including 130 method papers and 41 benchmarks, datasets, surveys, and analysis papers, we propose a representation-centered taxonomy that treats the form of the intermediate state as the organizing axis. We systematize the 130 methods into four categories: language-based, visual-spatial, latent-dynamic, and externalized reasoning, further divided into 13 subtypes tied to distinct regions of interests. Our synthesis shows that the open frontier of reasoning in driving agents lies in intermediate representations that can be grounded in the real world, coupled to real-time action, and verified under safety-critical systems. Project page: https://github.com/tangzhengxu/awesome-av-cot.

cs.CV↗

arXiv2026

ScaleSense: Cost-Intelligent Scaling Framework via Learned Resource Estimation in Alibaba AnalyticDB

Cloud-native serverless data warehouses achieve fine-grained elasticity by decoupling storage from compute, yet determining the optimal resource allocation for highly heterogeneous ad-hoc queries remains a formidable industrial challenge. Our analysis of production workloads in Alibaba AnalyticDB exposes a costly ``provisioning trap'': the fear of catastrophic resource depletion drives users to blindly over-provision resources, wasting immense monetary budgets without alleviating non-CPU bottlenecks (e.g., I/O saturation). To break this impasse, we propose ScaleSense, a proactive, query-level resource scaling framework. Specifically, it features a multi-faceted query encoder that jointly models plan topologies and hardware specifications. Crucially, a quantile-based resource predictor estimates multi-dimensional physical footprints, acting as a reliable safety net for optimal resource scaling. An auto-scaling controller then navigates the performance-cost Pareto frontier, dynamically tailoring allocations to specific business priorities without requiring model retraining. Evaluations on over 1.36 million production queries show that ScaleSense achieves state-of-the-art prediction accuracy with good prediction interval coverage. By achieving a 76.7% relative improvement in optimal resource configuration selection over the best baseline, this approach addresses the critical performance-cost trade-off while maintaining low-overhead inference latency, confirming its practical performance in production deployments. Under the performance-optimization policy, ScaleSense satisfies user-defined performance requirements while reducing monetary cost by up to 5.22x.

cs.DB↗

arXiv2026

A Dual-domain Refinement Network with FBP-based Jacobian Learning for Sparse-view Dual-Energy CT Material Decomposition

Dual-energy CT (DECT) exploits attenuation differences across different X-ray spectra to provide richer material information and has been widely used in medical imaging. While sparse-view acquisition can lower radiation exposure, it makes DECT material decomposition even more challenging, as the problem is nonlinear and ill-posed. Existing deep unrolling approaches generally do not explicitly incorporate the Jacobian operator induced by the nonlinear forward model, and their sparsity priors are still mainly built on conventional convolutions, which are insufficient for modeling global structural information. This study addresses the challenge of DECT multi-material decomposition in sparse-view settings by representing it as a sparse-regularized nonlinear least-squares problem. To solve it, we propose an iterative dual-domain refinement network (DECT-DRNet). In each iteration, the filtered back-projection (FBP)-based Jacobian approximation module is used first to generate an intermediate material decomposition result. Here, we characterize the forward process of material decomposition using a nonlinear operator, and then construct a theoretically grounded learnable approximation of the adjoint Jacobian operator by integrating the FBP algorithm with a U-Net into the backward process. In addition, to address the limitation of existing deep learning-based decomposition methods in globally suppressing noise and artifacts, we introduce a learnable sparse dual domain regularization term that incorporates Fourier convolutional residual blocks. This refinement block combines geometric feature extraction in the image domain with noise suppression in the frequency domain, allowing the model to capture both global and local features while maintaining structural details. DECT-DRNet demonstrates its ability to achieve more accurate material decomposition under sparse-view conditions.

cs.CV↗

arXiv2026

VoiceDesigner: Text-to-Voice Generation and Editing via Unified Diffusion Modeling and Data Augmentation

Recent breakthroughs in generative models have made text-to-voice generation (TTV) possible, enabling the synthesis of speech directly from textual voice descriptions. However, existing systems face two key challenges. First, they struggle to generate a diverse range of voices, spanning real-world human speakers and fictional characters. Second, they lack robust and flexible voice editing capabilities, such as voice cloning and the ability to modify attributes like emotion and tone. In this paper, we propose VoiceDesigner, a unified framework for text-to-voice generation and editing that supports diverse and controllable voice design. To tackle the above challenges, we propose solutions from two perspectives. First, we develop a hybrid data pipeline that leverages digital signal processing techniques and speech generation models to construct a diverse voice dataset covering both real-world and fictional voices. Second, we introduce a diffusion transformer with architectural improvements to better handle complex conditioning and enhance multi-task performance, enabling unified voice generation and editing. Through subjective and objective evaluations, VoiceDesigner achieves superior prompt alignment with both voice descriptions and editing instructions, while maintaining competitive perceptual quality and voice usability compared to state-of-the-art TTV models.

eess.AS↗

arXiv2026

ControlRadio: Prompt-Driven Controllable Diffusion for Cross-Modal Radio Map Generation

Radio maps describe how wireless signals propagate across space and are essential for wireless communication, sensing, and network planning. However, constructing accurate radio maps traditionally requires either dense measurements or computationally expensive physical simulations, which limits scalability and real-time deployment. Recent advances in generative artificial intelligence offer a promising alternative, but existing approaches lack fine-grained control and physical consistency when applied to real-world wireless environments. Here we present \textbf{ControlRadio}, a controllable generative framework that produces radio maps from natural-language descriptions and environmental layouts, including building structures and transmitter locations. Joint semantic and spatial conditioning enables interpretable, propagation-plausible generation, while a controlled latent prior and layout-aware conditioning improve stability and structural consistency. Extensive experiments demonstrate that ControlRadio achieves state-of-the-art accuracy and strong generalization across diverse urban scenarios, while reducing computation time by more than four orders of magnitude compared with conventional simulation-based methods. Such results suggest a new paradigm for scalable and controllable wireless environment modeling, with broad implications for next-generation communication systems and data-driven radio sensing.

cs.CV↗