Searcharxiv⌕ Search

arXiv subjects

Jun Cheng

Publications and source records attributed to Jun Cheng.

At least 19 recordsLinked to original sources
arXiv2026

IPM-FM: A Foundation Model with Consensus Feature Selection for Industrial Process Monitoring

Industrial process monitoring is fundamental to the safety and economic performance of modern process plants. Current practice remains a one-task-one-model paradigm that is label-inefficient and prone to degradation under operating drift. Foundation models have reshaped language, vision, and generic time-series forecasting, but it has not been adapted to industrial process monitoring. This setting poses domain-specific challenges, including safety-critical decisions and asymmetric sampling between process variables and laboratory measurements. We propose the industrial process monitoring foundation model (IPM-FM). It first learns general-purpose representations from unlabeled industrial process data through self-supervised pretraining, then adapts to specific monitoring tasks using a small amount of task-labeled data, and finally produces calibrated predictions through an uncertainty-aware prediction head. IPM-FM integrates a self-supervised Informer backbone with a multi-criteria consensus feature selector, a recursive lag-feature regression head, and a calibrated Monte Carlo dropout uncertainty module. On a seven-year hydrotreater dataset for diesel flash-point soft sensing, IPM-FM attains an RMSE of 2.99, $R^2$ of 0.50, and 97\% coverage of its 95\% predictive interval, outperforming the strongest classical and from-scratch sequence baselines by 8.3\% and 14.6\% in RMSE respectively, supporting the viability of a unified pretraining--adaptation framework for industrial process monitoring.

cs.LG↗
arXiv2026

Improving O-RADS Risk Stratification from Ultrasound Reports: A Comparative Evaluation of Hybrid versus End-to-End LLM Reasoning Strategies

Background: Automating clinical guideline-based decision-making with large language models (LLMs) remains challenging because of reliability, hallucination, and limited interpretability. We compared the performance of LLMs and reasoning strategies for automated Ovarian-Adnexal Reporting and Data System (O-RADS) classification from free-text pelvic ultrasound reports. Methods: In this retrospective study, consecutive patients with ovarian masses who underwent pelvic ultrasound were included. Eight LLMs were tested with three reasoning strategies: implicit-knowledge end-to-end, rule-informed end-to-end, and a feature-based hybrid architecture that decoupled feature extraction from rule-based classification. The reference standard was O-RADS categorization established by expert consensus. Results: A total of 310 women with 390 ovarian masses were evaluated. The feature-based hybrid architecture using Gemini 3.6 Flash demonstrated the best performance, achieving an accuracy of 99.2% (387 of 390) and almost perfect agreement with the reference standard (weighted kappa = 1.00; 95% CI: 0.99-1.00). Its performance surpassed that of original clinical reports (accuracy, 87.7% [342 of 390]; weighted kappa = 0.94; 95% CI: 0.91-0.96) and end-to-end LLM strategies (accuracy range, 65.6% [256 of 390] to 95.9% [374 of 390]). For structured feature extraction, Gemini 3.6 Flash demonstrated higher overall accuracy than Claude Fable 5 (98.9% vs 97.8%; P < 0.001). The hybrid architecture reduced misclassification errors and mitigated the overstaging tendency observed in original reports. Conclusion: The feature-based hybrid LLM architecture that separates clinical feature extraction from deterministic guideline execution enables highly accurate, reliable, and interpretable automated O-RADS classification, providing a promising approach for standardized, guideline-based clinical decision-making.

cs.AI↗
arXiv2026

ComBodied Agents: a New Paradigm of Human-Centric Agentic AI

After an older adult misses a medication dose, a software agent can send another reminder and an embodied agent can bring the medication. Yet neither explains whether the person forgot, is confused, has side effects, or deliberately refused, nor what support is appropriate. This reveals a structural gap in Agentic AI: Digital Agents primarily transform software states, while Embodied Agents transform physical states; neither makes a person's evolving state and agency the primary object of modeling, intervention, and evaluation. We introduce Combodied Agents, a human-centered paradigm that perceives, models, predicts, and supports individual human-state trajectories over time, using software tools, sensors, wearables, robots, and human services as action channels rather than end goals. We unify fragmented capabilities across personal assistants, health agents, AI companions, and adaptive human--AI systems into a closed loop: event-based multimodal perception reconstructs meaningful personal events; longitudinal, correctable memory provides temporal context; Personal World Models estimate future personal states and outcomes under alternative decisions and interventions; and an admissible intervention policy selects proportionate support under consent, uncertainty, safety, reversibility, and user control. Feedback from the person and environment updates the loop. Rather than requiring an exhaustive Human Digital Twin, the framework uses purpose-bounded, uncertainty-aware, user-correctable representations. We organize the design space by human-state targets, relational contexts, and agent roles, and propose scenario-centered evaluation, agency-preservation metrics, benchmark requirements, edge-native personal models, and governance directions. Combodied Agents shift Agentic AI from external task completion toward sustained human benefit.

cs.AI↗
arXiv2026

3D Molecular Representation Learning for Organic Mixtures: Viscosity and Density Prediction

The viscosity and density of organic mixtures are essential properties for designing lubricants, solvents, and heat transfer fluids. In engineering practice, formulating a functional fluid requires understanding how these properties change with composition and temperature. However, exhaustive experimental characterization across the full parameter space is impractical due to the vast number of possible species and combinations. Here we introduce a mixture-aware 3D molecular representation learning strategy, built upon a pre-trained molecular encoder, that jointly encodes component structures, mole fractions, and temperature to achieve accurate predictions for organic mixtures. Fine-tuning on publicly available datasets covering a wide range of binary organic mixtures yields test-set R2 values of 0.973 for dynamic viscosity and 0.996 for density, significantly outperforming traditional machine learning baselines. Beyond this overall accuracy, the model captures non-monotonic viscosity changes upon mixing, surpassing simple linear or logarithmic mixing rules. The architecture is extendable to ternary and multicomponent mixtures, as verified via preliminary experiments. Using this model, we quantitatively analyze how molecular structure-branching, cycloalkane, and aromatic rings-affects viscosity-temperature behavior, which benefits the design of lubricants with superior viscosity-temperature performance. Altogether, this work provides a practical, data-driven tool for mixture property prediction, accelerating the rational formulation of functional fluids in chemical engineering.

physics.chem-ph↗
arXiv2026

Uni-XAS: Alignment-Driven Bidirectional Multimodal Learning for X-ray Absorption Spectroscopy

X-ray absorption spectroscopy (XAS) is a key technique for probing local atomic environments, yet learning based modeling must bridge two heterogeneous modalities: 1D continuous spectra and 3D atomic structures. Existing approaches typically decouple forward spectrum prediction and inverse structure inference into separate regression tasks, hindering shared representation learning. Moreover, severe permutation ambiguity among identical atoms often limits inverse modeling to coarse structure descriptors rather than explicit 3D structure generation. In this work, we present Uni-XAS, a unified benchmark and learning framework that reframes bidirectional XAS modeling as a cross-modal alignment and conditional generation problem. We first propose XASLip, an alignment recipe coupling a physics-aware spectral encoder with an absorberaware manifold optimization strategy to resolve fine-grained intra-element coordination variations. Building upon this shared latent space, we formulate forward prediction as anchored absolute-spectrum generation via retrieval-augmented decoding, effectively preventing physical scale collapse and energy drift. For the inherently ill-posed inverse problem, we introduce Permutation-Rectified Flow Matching, which integrates type-wise optimal transport into a continuous generative flow to provide a principled solution to ligand permutation ambiguity without relying on heavy high-order equivariant architectures. Evaluated on a largescale standardized benchmark of 328,839 structure-spectrum pairs, Uni-XAS demonstrates strong performance in cross-modal retrieval, accurate absolute-spectrum prediction, and composition-conditional 3D structure generation, establishing a scalable, reproducible, and protocol-consistent foundation for multimodal learning and standardized evaluation in scientific spectroscopy.

cond-mat.mtrl-sci↗
arXiv2026

Coordinate Independence of the Schwarzschild Black Hole Accretion Vlasov Gas Model

This paper presents a detailed study of the coordinate dependence of Vlasov gas accretion onto a Schwarzschild black hole. Asymptotic results at infinity and near horizon are obtained via Taylor expansions for three different statistical distributions within the framework of the most general stationary spherically symmetric spacetime. Our findings demonstrate that the particle number density, energy density, radial and tangential pressures, and accretion rates are independent of the coordinate choice, even though individual components such as the particle current density and the stress-energy tensor explicitly depend on the coordinate system. Consequently, the accretion theory can be formulated without reference to any particular coordinate system. We also show that the mean energy of the accreted particles is $m_0+k_BT$, lower than the mean energy $m_0+\frac{3}{2}k_BT$ of the Maxwell-Boltzmann system in the classical limit. And the specific entropy of the accreted particles is lower than the global average by $\frac{3}{2}k_B$. This is because particles of lower energy are more easily accreted, while particles of higher energy are more readily scattered. We also present numerical results at finite radii for the relevant physical quantities.

gr-qc↗
arXiv2026

Enhanced ODMA for Massive Sparse Access: Hybrid Data-Frozen Bit Transmission and Fixed-Point Analysis over Block Fading Channels

This paper proposes a novel on-off division multiple access (ODMA) transmission scheme that enables efficient joint multi-user channel estimation and iterative decoding by inserting a small number of frozen symbols into the codewords. Functionally analogous to pilots, these symbols are sparsely distributed within the codeword. Unlike conventional pilot-based methods, our approach requires only a minimal number of frozen symbols (e.g., $5\sim20$ symbols per user in a 300-user system), which serve dual purposes as both estimation references and decoding aids. By employing low-complexity single-user channel estimation and decoding, combined with simple iterative interference cancellation, the scheme achieves performance equivalent to that with perfectly known user channels, even when accounting for the additional energy and bandwidth costs of the frozen symbols. Furthermore, for the large-scale ODMA sparse multiple access system, this paper proposes a fixed-point analysis method, which can accurately estimate the iterative convergence performance over multi-user block fading channels by only leveraging the decoding functions under the single-user AWGN channel. This method is applicable to performance analysis for arbitrary code lengths, code rates, and decoding algorithms. It eliminates the need for extensive Monte Carlo simulation time, and provides an efficient tool for the design of multi-user codes.

cs.IT↗
arXiv2026

Full-data accuracy with fewer labels for training and fine-tuning machine-learning force fields

Machine-learning force fields (MLFFs) are reliable only near their training distribution, making efficient construction of diverse training sets a major bottleneck for both train-from-scratch and foundation fine-tuning workflows. Active learning can reduce this cost, but standard model-committee uncertainty is impractical for foundation MLFFs because each committee member requires a separate fine-tuning run. We present an active-learning workflow based on last-layer-projection regression (LLPR), a forward-pass-cheap per-configuration uncertainty estimator. Across molecular, condensed-phase, and electrolyte systems, LLPR identifies compact, high-value training sets that recover full-data accuracy using only a small fraction of electronic-structure labels. In foundation-model fine-tuning, LLPR-selected configurations reach the full-pool fine-tuning ceiling with substantially fewer labels than random selection. In iterative electrolyte fine-tuning, LLPR detects unphysical local coordination before DFT labelling, provides an absolute force-error threshold, and enables automatic termination of the learning loop. The resulting models reproduce reference density and ion-coordination structure, providing a scalable uncertainty-quantification strategy across MLFF training regimes.

physics.chem-ph↗
arXiv2026

Ai2-Kit: Streamlining AI-Accelerated Ab Initio Workflows for Complex Chemical Systems

Molecular simulations of complex chemical systems, such as catalysis, electrochemistry, and energy storage, often need to capture the interplay of effects such as electronic structure, finite-temperature fluctuations, and electric-field response. Such complexity is difficult to address with traditional ab initio calculations, which are limited by the time and length scales they can reach. AI-accelerated ab initio (AI2) methods use machine learning potentials trained on first-principles data to replace expensive electronic-structure calculations, extending ab initio accuracy to these regimes, but their routine application requires reliable workflows that connect first-principles calculations, model training, molecular dynamics, enhanced sampling, trajectory analysis, and HPC orchestration. Here we present ai2-kit, a software toolkit for developing accessible, reproducible, and extensible AI2 workflows. ai2-kit provides high-semantic-density command-line interfaces and Python APIs for structure and dataset conversion, batch task generation, active-learning screening, job orchestration, and workflow recovery. We demonstrate ai2-kit in four representative applications: active-learning-based machine learning potential construction, free-energy perturbation for redox and acid-base processes, electrochemical machine learning potentials for electrified interfaces, and spectroscopies from machine learning molecular dynamics. ai2-kit also provides AI-agent skills that help users adapt these use cases into customized workflows for their own chemical systems and computational software stacks. Together, ai2-kit helps turn AI2 methods from bespoke computational protocols into reusable and extensible workflows for complex chemical systems, from model construction to property prediction.

physics.chem-ph↗
arXiv2026

MARVEL: Universal Murray's Law-informed Vessel Tree Segmentation and Topology Estimation

Vascular circulation follows fundamental biophysical principles that optimize mass transport and metabolic energy expenditure, which can be effectively modeled by Murray's law. However, contemporary deep learning methods for vascular segmentation often neglect these biophysical constraints. This leads to physiologically implausible branching and misclassification vascular trees, rendering. These automated segmentation results are unreliable unreliable for downstream clinical tasks such as blood flow simulation or disease quantification. In this paper, we introduce MARVEL (Universal MurrAy's law-infoRmed Vessel sEgmentation and topoLogy estimation), a backbone-agnostic framework that integrates biophysical priors into vascular tree extraction. MARVEL combines per-pixel supervision with explicit radius predictions to enforce local bifurcation constraints derived from an empirical width-exponent mapping. We implement these constraints as differentiable regularizers during training to guide models toward physiologically consistent reconstructions. We evaluate MARVEL on eight public datasets across multiple vascular modalities and segmentation backbones. Results demonstrate MARVEL's superior performance in segmentation accuracy, topological consistency, and physiological plausibility. By converting segmented masks into graph-based hemodynamic simulations, we demonstrate that MARVEL preserves the subtle pathological narrowing and topological connectivity required to distinguish hypertensive from normotensive eyes. Results show that MARVEL significantly improves the classification of hypertension via arteriovenous pressure differences in the eye (p < 0.001), outperforming baseline models in both topological consistency and clinical predictive value.

cs.CV↗
arXiv2026

Tunneling spectra of $\mathrm{TaO}_x$ junctions for van der Waals superconductors

Tunneling spectroscopy and its evolution are crucial for elucidating the intricate electronic structure and emergent phenomena in quantum materials.Nevertheless, high-quality measurements -- specifically those tracking evolution across temperature and external fields -- remain a formidable challenge. We have fabricated a high-quality $\mathrm{TaO}_x$-based planar tunneling junction by using magnetron sputtering for van der Waals (vdW) superconductors. Using the vdW superconductor $\mathrm{Bi}_2\mathrm{Sr}_2\mathrm{CaCu}_2\mathrm{O}_{8+\delta}$ (Bi2212) as a benchmark, this platform yields high-quality tunneling spectra, reproducing the electronic signatures obtained from scanning tunneling spectra acquired from atomically clean surfaces under ultra-high vacuum conditions. This architecture enables high-precision spectroscopy across extensive temperature and magnetic field ranges, offering a universal strategy for probing the electronic structures of diverse two-dimensional systems and facilitating future explorations of material properties.

cond-mat.supr-con↗
arXiv2026

One-hot Coding-based URA with RFFI-Enabled Message Authentication

Unsourced random access (URA) has emerged as a promising paradigm for enabling massive connectivity in Internet-of-Things (IoT) networks. However, since URA transmissions do not contain device identifiers, the receiver may not associate decoded messages with their originating devices, introducing a security vulnerability: forged messages may be decoded as legitimate. To address this problem, this paper proposes a one-hot coding (OHC)-based URA framework that enables message authentication while preserving the unsourced transmission principle. Specifically, distinct messages are mapped onto orthogonal channel uses via an OHC-based common codebook and transmitted using on-off keying modulation. The resulting orthogonal channel structure enables radio-frequency fingerprint identification to authenticate received signals by exploiting device-specific hardware impairments, thereby authenticating decoded messages without introducing an additional authentication payload. Analytical expressions for the per-user probability of error and the probability of successful spoofing are derived. Numerical results demonstrate that the proposed scheme enables secure URA transmission while maintaining reliable communication performance in ultra-short-payload IoT scenarios.

eess.SP↗
arXiv2026

Degradation-Aware Adaptive Context Gating for Unified Image Restoration

Unified image restoration using a single model often faces task interference due to diverse degradations. To address this, we propose DACG-IR (Degradation-Aware Adaptive Context Gating), which enables explicit perception of degradation characteristics to dynamically modulate feature representations. Our method constructs degradation-aware contextual representations from the input to modulate attention distribution, frequency-domain features, and feature aggregation. Specifically, a lightweight multi-scale degradation-aware module extracts coarse degradation information and generates layer-wise prompts. These prompts guide attention temperature and output gating in encoder and decoder blocks for adaptive feature extraction. Additionally, a spatial-channel dual-gated adaptive fusion mechanism refines encoder features, suppressing noise propagation from shallow to deep layers. This design effectively suppresses degradation-induced noise while preserving informative structures. Experiments show DACG-IR outperforms state-of-the-art methods in single-task, all-in-one, adverse weather removal, and composite degradation settings. Code: https://github.com/HlHomes/DACG-IR-code

cs.CV↗
arXiv2026

Circular Phase Representation and Geometry-Aware Optimization for Ptychographic Image Reconstruction

Traditional iterative reconstruction methods are accurate but computationally expensive, limiting their use in high-throughput and real-time ptychography. Recent deep learning approaches improve speed, but often predict phase as a Euclidean scalar despite its $2\pi$ periodicity, which can introduce wrapping artifacts, discontinuities at $\pm\pi$, and a mismatch between the loss and the underlying signal geometry. We present a deep learning framework for ptychographic reconstruction that models phase on the unit circle using cosine and sine components. Phase error is optimized with a differentiable geodesic loss, which avoids branch-cut discontinuities and provides bounded gradients. The network further incorporates saturation-aware dual-gain input scaling, parallel encoder branches, and three decoders for amplitude, cosine, and sine prediction, together with a composite loss that promotes circular consistency and structural fidelity. Experiments on synthetic and experimental datasets show consistent improvements in both amplitude and phase reconstruction over existing deep learning methods. Frequency-domain analysis further shows better preservation of mid- and high-frequency phase content. The proposed method also provides substantial speedup over iterative solvers while maintaining physically consistent reconstructions.

eess.IV↗
arXiv2026

NOSE: Neural Olfactory-Semantic Embedding with Tri-Modal Orthogonal Contrastive Learning

Olfaction lies at the intersection of chemical structure, neural encoding, and linguistic perception, yet existing representation methods fail to fully capture this pathway. Current approaches typically model only isolated segments of the olfactory pathway, overlooking the complete chain from molecule to receptors to linguistic descriptions. Such fragmentation yields learned embeddings that lack both biological grounding and semantic interpretability. We propose NOSE (Neural Olfactory-Semantic Embedding), a representation learning framework that aligns three modalities along the olfactory pathway: molecular structure, receptor sequence, and natural language description. Rather than simply fusing these signals, we decouple their contributions via orthogonal constraints, preserving the unique encoded information of each modality. To address the sparsity of olfactory language, we introduce a weak positive sample strategy to calibrate semantic similarity, preventing erroneous repulsion of similar odors in the feature space. Extensive experiments demonstrate that NOSE achieves state-of-the-art (SOTA) performance and excellent zero-shot generalization, confirming the strong alignment between its representation space and human olfactory intuition.Code and data are available at https://github.com/Xianyusyy/NOSE

cs.CL↗
arXiv2026

Solving the inverse problem of X-ray absorption spectroscopy via physics-informed deep learning

Resolving transient atomic configurations in non-crystalline or dynamic environments remains a fundamental bottleneck in the physical sciences. While X-ray absorption spectroscopy (XAS) is a premier probe of local structure, inverting spectra into structural descriptors is a notoriously ill-posed problem due to inherent many-to-one mapping. Here, we present the Spectral Pattern Translator (SPT), a physics-informed deep learning framework that establishes a robust bridge between large-scale theoretical datasets and experimental reality. Our strategy exploits the Fourier duality between spectral energy oscillations and spatial scattering paths to overcome the "simulation-to-experiment" gap. By decomposing spectra into frequency domains, SPT effectively isolates robust structural coordination signals from the destabilizing noise inherent in experimental data. Trained on a massive library of diverse atomic environments, this approach achieves state-of-the-art accuracy in resolving continuous phase transitions in battery cathodes and deciphering local order in amorphous materials. With millisecond-scale latency, SPT removes the primary computational barrier to autonomous materials discovery, establishing a robust, noise-resilient engine for closed-loop robotic chemistry.

cond-mat.mtrl-sci↗
arXiv2026

Spectral-Geometric Neural Fields for Pose-Free LiDAR View Synthesis

Neural Radiance Fields (NeRF) have shown remarkable success in image novel view synthesis (NVS), inspiring extensions to LiDAR NVS. However, most methods heavily rely on accurate camera poses for scene reconstruction. The sparsity and textureless nature of LiDAR data also present distinct challenges, leading to geometric holes and discontinuous surfaces. To address these issues, we propose SG-NLF, a pose-free LiDAR NeRF framework that integrates spectral information with geometric consistency. Specifically, we design a hybrid representation based on spectral priors to reconstruct smooth geometry. For pose optimization, we construct a confidence-aware graph based on feature compatibility to achieve global alignment. In addition, an adversarial learning strategy is introduced to enforce cross-frame consistency, thereby enhancing reconstruction quality. Comprehensive experiments demonstrate the effectiveness of our framework, especially in challenging low-frequency scenarios. Compared to previous state-of-the-art methods, SG-NLF improves reconstruction quality and pose accuracy by over 35.8% and 68.8%. Our work can provide a novel perspective for LiDAR view synthesis.

cs.CV↗
arXiv2026

PhaForce: Phase-Scheduled Visual-Force Policy Learning with Slow Planning and Fast Correction for Contact-Rich Manipulation

Contact-rich manipulation requires not only vision-dominant task semantics but also closed-loop reactions to force/torque (F/T) transients. Yet, generative visuomotor policies are typically constrained to low-frequency updates due to inference latency and action chunking, underutilizing F/T for control-rate feedback. Furthermore, existing force-aware methods often inject force continuously and indiscriminately, lacking an explicit mechanism to schedule when / how much / where to apply force across different task phases. We propose PhaForce, a phase-scheduled visual--force policy that coordinates low-rate chunk-level planning and high-rate residual correction via a unified contact/phase schedule. PhaForce comprises (i) a contact-aware phase predictor (CAP) that estimates contact probability and phase belief, (ii) a Slow diffusion planner that performs dual-gated visual--force fusion with orthogonal residual injection to preserve vision semantics while conditioning on force, and (iii) a Fast corrector that applies control-rate phase-routed residuals in interpretable corrective subspaces for within-chunk micro-adjustments. Across multiple real-robot contact-rich tasks, PhaForce achieves an average success rate of 86% (+40 pp over baselines), while also substantially improving contact quality by regulating interaction forces and exhibiting robust adaptability to OOD geometric shifts.

cs.RO↗