Searcharxiv⌕ Search

arXiv subjects

Yuhao Chen

Publications and source records attributed to Yuhao Chen.

At least 19 recordsLinked to original sources
arXiv2026

Fitting Conventional and Neural Exchange-Correlation Functionals to Accurate Densities without Self-Consistent Fields

Exchange-correlation functionals are usually fitted with energy-only losses, so better energies often bring worse densities. We introduce a workflow that predicts in closed form, from one Fock build on inverse Kohn-Sham orbitals of any accurate density, a functional's self-consistent density, and fits it to that density with no self-consistent-field solution. Fitted convexly, a CF22D-form hybrid beats CF22D and MN15 on unseen energies, dipole moments, and unfitted densities (6.8- and 22-fold lower density error). A neural meta-GGA fitted to 6 energies and 11 densities beats SCAN in both.

physics.chem-ph↗
arXiv2026

UT-HMR: Uncertainty-Guided Temporal Modeling for Robust 3D Human Mesh Recovery

Monocular 3D human mesh recovery (HMR) remains unreliable under weak observations caused by long camera-to-person distances, occlusion, and motion blur. In challenging scenarios such as sports, these factors make the problem highly ambiguous, as different 3D poses may yield similar 2D projections. This ambiguity is especially severe for one-stage HMR, which recovers all meshes directly from the full image without person crops or external 2D cues. To handle this ambiguity, we present UT-HMR ( Uncertainty-guided Temporal HMR), to our knowledge the first one-stage HMR framework to capture correlated pose uncertainty across joints. Each person query predicts a Gaussian distribution over the full SMPL pose with a compact low-rank-plus-diagonal covariance, which is converted into joint-wise precision weights for heteroscedastic supervision while keeping inference deterministic. Temporal early fusion further injects neighboring-frame evidence before pose decoding, supplying the visual evidence needed to resolve the ambiguity that the uncertainty captures. With a lightweight ViT-B backbone, UT-HMR achieves state-of-the-art results on VIP-HockeyPose-v2 (ice hockey) and WorldPose (soccer), reducing MPJPE, PA-MPJPE, and PVE by up to 4.7%, 3.8%, and 5.6%, and remains competitive on 3DPW against methods with larger backbones and higher input resolutions.

cs.CV↗
arXiv2026

Environment-Adaptive Dynamic KV Cache for Multi-Path Parameter Prediction

Multipath parameter prediction provides a structured and physically interpretable representation of wireless propagation for environment-aware communication systems. However, existing Transformer-based methods repeatedly encode the same environment when predicting multiple BS--UE links, resulting in redundant computation. To address this issue, we propose an environment-conditioned dynamic KV caching framework for multipath parameter prediction. Specifically, we decouple environment-dependent computation from BS--UE geometry, enabling environment-side representations and intermediate KV states to be reused across multiple link queries while only the geometry branch is updated. Building on this reusable cache, we further introduce a dynamic compression mechanism that uses environment-side attention scores to retain informative tokens and adaptively allocates a fixed cache budget across Transformer layers. Experiments on the WAIR-D dataset show that the proposed method reduces KV memory from 0.2500 MB to 0.1406 MB, while the delay NMSE increases only slightly from 0.0673 to 0.0678. Code is available at https://github.com/LZhaoyang/kv.

eess.SP↗
arXiv2026

STAIF: A Stage-wise Optimization for Complex Instruction Following

Following complex instructions with multiple explicit constraints remains a fundamental challenge for large language models (LLMs). Existing alignment methods, such as DPO, optimize holistic reward signals that often underemphasize strict satisfaction of individual constraints, particularly under out-of-distribution or multi-constraint settings. In this paper, we propose STAIF, a stage-wise optimization framework that decouples the alignment of subjective (soft) constraints from the optimization of objectively verifiable (hard) constraints. Stage 1 applies preference optimization with multiple negative samples to sharpen sensitivity to soft constraints, while Stage 2 applies Reinforcement Learning with Verifiable Rewards (RLVR) to enforce strict compliance with hard constraints. To support this method, we construct STAINSTRUCT, a high-quality bilingual (English, Chinese) dataset of approximately 31,000 complex multi-constraint instructions. Extensive analyses validate the design of STAIF and show state-of-the-art performance on representative benchmarks against strong baselines, as well as genuine generalization.

cs.AI↗
arXiv2026

DPPM: Dual-Path Parametric Memory for Personalized Language Models

Long-term personalization requires language models to use interaction history to track users' preferences across sessions. Parametric memory encodes this interaction history into model parameters or adapters, reducing the need to include it in the inference context. However, independent context compilation leaves cross-session integration unspecified, while recurrent updates can attenuate earlier evidence. To address these challenges, we propose Dual-Path Parametric Memory (DPPM). Its Evidence path directly pools representations of the interaction history to preserve earlier evidence, while its Delta path sequentially updates an associative state to capture changes. Fusing both outputs produces history-conditioned LoRA adapters that combine evidence accumulation with ordered revision. Across multiple backbones, DPPM outperforms the evaluated baselines, achieving 54.22% on PersonaMem-v2 and 86.79% on PrefEval. These results suggest that DPPM provides a simple and effective design choice for cross-session personalized parametric memory.

cs.CL↗
arXiv2026

Subspace Consensus

This paper examines the subspace consensus problem for matrix-weighted multi-agent networks. In this setting, each agent holds a high-dimensional state vector in $\mathbb{R}^{d}$, and matrix-valued edge weights characterize inter-agent interactions across different dimensions. Unlike the classical scalar-weighted consensus problem, where all dimensions of agents' states must converge to respective common values, this paper examines scenarios in which agents must agree along certain dimensions while preserving prescribed relative configurations in the remaining ones. To address this need, we introduce the concept of subspace consensus for a matrix-weighted network, which is said to achieve consensus on a subspace $\mathbb{V}\subseteq\mathbb{R}^{d}$ if the projections of all pairwise state differences onto $\mathbb{V}$ asymptotically vanish. This formulation subsumes the conventional consensus problem as the special case $\mathbb{V}=\mathbb{R}^{d}$. We first derive a purely algebraic necessary and sufficient condition for subspace consensus by characterizing the interplay between the null spaces of the matrix-valued Laplacian and those of the individual matrix-valued edge weights. Then, we propose the limiting constraint network as a means to analyze the steady state of matrix-weighted networks. We establish the relationship between the steady state of a matrix-weighted network and the null space of the matrix-valued Laplacian of the associated limiting constraint network. Under mild assumptions, we further show that a matrix-weighted network achieves subspace consensus on $\mathbb{V}$ if and only if the associated limiting constraint network contains a $\mathbb{V}$-spanning tree. We further show that a matrix-weighted network achieves consensus if and only if the associated limiting constraint network contains a positive spanning tree, as a special case with $\mathbb{V}=\mathbb{R}^{d}$.

eess.SY↗
arXiv2026

Pushing the accuracy of on-top functionals with agent-driven supervised learning

Multiconfiguration pair-density functional theory (MC-PDFT) combines multiconfigurational reference wave functions with on-top functionals to provide an efficient electronic-structure method for strongly correlated molecular systems. We developed a large language model-based agent for functional development and combined multiple databases into the large, chemically diverse MCDDB26 database. We then used MCDDB26 to develop two on-top functionals, MC26 and COF26. MC26 retains an existing analytical form but substantially improves performance by refitting its linear parameters with optimised weights for the training datasets. Building on this result, we developed COF26, a general hybrid on-top functional with a new analytical form. COF26 provides greater overall accuracy for multireference chemistry than existing on-top functionals. Applications across MCDDB26 show that COF26 accommodates both multiconfigurational and single-configuration densities. It therefore performs well for barrier heights, isomerisation energies, thermochemistry, non-covalent interactions, radical and non-radical chemistry, single-reference and multireference systems, and transition-metal chemistry.

physics.chem-ph↗
arXiv2026

SPARC: SuperPixel-Aware Region Contrastive Learning for Self-Supervised Dense Prediction

Self-supervised learning (SSL) has become an effective approach for learning visual representations without manual annotations. Among SSL approaches, contrastive learning has been widely used for visual representation learning. However, existing contrastive SSL methods have focused primarily on image-level or pixel-level representation learning, while region-level representation learning remains less explored. We propose SPARC, a region-level contrastive learning framework that leverages superpixels to establish explicit correspondence between augmented image views. SPARC introduces a region contrastive branch that performs superpixel-based feature pooling and optimizes a region-level contrastive objective jointly with a global image-level objective. Under identical settings, SPARC consistently outperforms previous methods such as MoCo-v2 and DenseCL, achieving improvements of up to +9.79 mIoU for semantic segmentation and +4.88 AP for object detection. Ablation studies further demonstrate that region-level objectives produce the strongest performance. Thus, region-level contrastive learning is an effective approach for improving self-supervised visual pretraining for dense prediction tasks. Code repository can be accessed at https://github.com/xRIPEIx/SPARC.

cs.CV↗
arXiv2026

Plasmoid-Trapped Condensation Associated with a Transient Thermal-Instability-Like Process in Chromospheric Magnetic Reconnection

We perform high-resolution 2.5D single-fluid magnetohydrodynamic (MHD) simulations to explore the development of cool-dense structures within plasmoids during chromospheric magnetic reconnection. The model incorporates temperature-dependent ionization degrees of hydrogen and helium, resulting in improved diffusivities and viscosity, as well as adequate radiative cooling. The numerical results show that plasma is significantly hot with temperatures reaching several tens of thousands of Kelvin in newly developed plasmoids in the low initial plasma-$β$ cases, $β_0$ = 0.05. Later, the localized accumulation of plasma and decreasing temperature result in an explosive much stronger radiative cooling process. Radiative cooling peaks over the temperature range 8000 K$-$20,000 K. The co-spatial density enhancement, substantial radiative cooling, shorter radiative cooling timescales, and a significant temperature decrease support the existence of thermal-instability-like condensation within the plasmoids. Similar cool-dense structures are identified in all low-$β_0$ cases at different chromospheric altitudes, suggesting that this occurrence is not limited to a single chromospheric layer. In contrast, in the high initial plasma-$β$, $β_0$ = 0.5 case, the maximum temperature inside the dense plasmoids is only about 8000 K. Therefore, the radiative cooling has never entered into the explosive increasing stage and the thermal-instability-like condensation does not happen. These findings reveal that plasmoid-trapped condensation may develop in chromospheric reconnection once radiative losses become self-amplifying under low-$β_0$ conditions.

astro-ph.SR↗
arXiv2026

TLive-Omni: An Omni-Modal Understanding Model for E-Commerce Live Streaming

E-commerce live streaming requires omni-modal understanding of noisy, temporally extended streams, where product facts are distributed across speech, video frames, product images, overlaid text, and user queries. We present TLive-Omni, an omni-modal understanding model tailored to live-commerce scenarios. It maps image, video, audio, and text inputs into a unified representation space. For long-form live streaming analysis, we introduce Per-vGrid, a timestamped token organization that groups each video grid with its temporally corresponding audio within explicit boundary tokens to facilitate temporal alignment. We design a three-stage supervised training recipe that progressively develops live-commerce understanding, from omni-modal perception to instruction-following responses. We then propose Faithful-RFT, a reinforcement fine-tuning stage that further improves answer faithfulness and expression quality while meeting real-time demands, scoring final responses directly with task-verifiable feedback rather than optimizing for reasoning-style exploration during rollout. Moreover, TLive-Omni is supported by a scenario-oriented atomic capability taxonomy and a compact data production engine that converts live-commerce audio, image, and video streams into training signals for speech recognition, speaker analysis, product visual grounding, text recognition, temporal grounding, video dense caption, and omni-modal QA, etc. For scalable training, a synchronized length-grouped sampler reduces padding while preserving comparable workloads across workers, while a lightweight dynamic sampling strategy regenerates rollout groups with near-zero reward variance to maintain meaningful relative advantages for GRPO. Experiments on e-commerce live streaming benchmarks demonstrate strong performance across live-commerce domain tasks, together with excellent generalization on general benchmarks.

cs.AI↗
arXiv2026

PLeDO: Pain Level Detection for Osteoarthritis from EMR Data

Osteoarthritis (OA) is a progressive chronic joint disease resulting in a breakdown of articular cartilage and bone when damaged joint tissues are not able to normally repair themselves. The aim of this pilot research study is to understand the pain severity for OA from patients' primary care Electronic Medical Records (EMR), both from the structured medical data and the unstructured chart note data using information extraction, natural language processing and machine learning techniques. We propose SPaDe, a Synonym-based Pain level Detection tool to categorize patients into having mild or moderate-to-severe pain to understand diagnosis and treatment methods based on only the pain related expressions in the unstructured chart note. Expressions are subjective, objective, and influenced by cultural background and demography which poses a difficult challenge. Therefore, we improve the model by incorporating the medication information from the structured EMR data and pain scale related information from the chart note to propose an integrated pain level detection tool for OA called PLeDO. With the help of human labeled gold standard data, we demonstrate that both SPaDe and PLeDO can detect mild and moderate-to-severe pain from the EMR data to analyze and potentially improve the quality of care in primary care setting.

cs.AI↗
arXiv2026

Divergent-Convergent Reasoning: Scaling Test-Time Compute through Structured Solution Synthesis

Test-time compute can substantially improve Large Language Model (LLM) reasoning performance, yet how and when additional compute helps remains poorly understood. We study Divergent-Convergent Reasoning (DCR), a simple two-phase primitive consisting of an exploration phase that generates multiple candidate solutions followed by a convergent reconciliation phase. We present three core results. First, we show that even a single reconciliation step can reliably amplify correct minority reports: across datasets, DCR often recovers the correct answer when correct exploration outputs are in the minority, a regime where majority voting fails. Second, we introduce recursive DCR, an autoregressive reconciliation system that iteratively analyzes disagreements and allocates additional test-time compute. Recursive DCR achieves higher accuracy than fixed-compute baselines-reaching 93.3% on AIME 2024 and 92.0% on AIME 2025-while using roughly 27% less compute on average, demonstrating that attentive resource allocation is superior to uniform scaling. Third, we analyze disagreement among exploration outputs via a simple, training-free dispersion metric. Dispersion reveals a structured relationship between disagreement and test-time gains: in regimes where DCR is effective, higher disagreement among exploration outputs is associated with larger accuracy improvements from reconciliation. Together, these results show that disagreement, often viewed as noise, can be systematically exploited to improve test-time reasoning and reveal emerging scaling laws for agentic LLM systems.

cs.AI↗
arXiv2026

CrossAudit: A Git-Native, Cross-Vendor Audit Loop for Agentic Science

An AI scientist should not grade its own homework. Yet in the systems we examined, the agent that reviews the work usually comes from the same model family as the agent that produced it, or at least from the same vendor. Model evaluators are known to favour their own generations. Whether models trained alike also share blind spots is a conjecture, not a settled finding, but if they do, the reviewer inherits the author's. The record of what was flagged and what was waved through often sits in platform logs that nobody outside can replay. We present CrossAudit, a protocol for supervising autonomous research pipelines. It rests on three commitments. Each increment of work is audited by an agent from a different vendor against a rulebook a human wrote and versioned. Reports, verdicts, disputes and rulings are git commits, so the supervision history can be re-read and cited; raw model exchanges are not yet part of that record. Scripted checks run before any model does. Advisory judgement never gates the pipeline: a model blocks only by citing a rule, and no model may waive a deterministic failure. Blockers that survive a bounded number of revision rounds go to a person. We state the protocol as eight invariants. We describe a reference implementation built from GitHub Actions and a few hundred lines of Python, and report a live deployment of a closely related variant in a computational-chemistry pipeline. We also ran a seeded-defect trial (30 increments, 43 seeded defects, one run per configuration). A cross-vendor audit of our own repository then voided its blinding. We adopt that audit's findings and report the corrected results. The trial shows that two vendors read the same rulebook differently. It does not show that either is better. The strongest evidence here is the committed, uncontrolled record of cross-vendor audits of this paper itself.

cs.AI↗
arXiv2026

TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework

Retrieval-Augmented Generation (RAG) utilizes external knowledge to augment Large Language Models' (LLMs) reliability. For flexibility, agentic RAG employs autonomous, multi-round retrieval and reasoning to resolve queries. Although recent agentic RAG has improved via reinforcement learning, they often incur substantial token overhead from search and reasoning processes. This trade-off prioritizes accuracy over efficiency. To address this issue, this work proposes TeaRAG, a token-efficient agentic RAG framework capable of compressing both retrieval content and reasoning steps. 1) First, the retrieved content is compressed by augmenting chunk-based semantic retrieval with a graph retrieval using concise triplets. A knowledge association graph is then built from semantic similarity and co-occurrence. Finally, Personalized PageRank is leveraged to highlight key knowledge within this graph, reducing the number of tokens per retrieval. 2) Besides, to reduce reasoning steps, Iterative Process-aware Direct Preference Optimization (IP-DPO) is proposed. Specifically, our reward function evaluates the knowledge sufficiency by a knowledge matching mechanism, while penalizing excessive reasoning steps. This design can produce high-quality preference-pair datasets, supporting iterative DPO to improve reasoning conciseness. Across six datasets, TeaRAG improves the average Exact Match by 4% and 2% while reducing output tokens by 61% and 59% on Llama3-8B-Instruct and Qwen2.5-14B-Instruct, respectively. Code is available at https://github.com/Applied-Machine-Learning-Lab/TeaRAG.

cs.IR↗
arXiv2026

A Simulation Framework with Guaranteed Error Bounds for Safety and Fuel-Efficiency Analysis of Vehicle Platoons

Vehicle platooning is an important technology in modern transportation systems, offering significant improvements in highway traffic efficiency and fuel economy. Achieving coordinated behavior among vehicles in a platoon depends on wireless communication. However, packet losses in wireless communication can create critical safety issues when they occur together with sudden braking. In this paper, we propose a rigorous simulation-based method for studying such safety issues by analyzing the minimum inter-vehicle distance over time across control parameters that guarantee string stability. In particular, our method computes the exact distance at simulation time instants and guarantees that the change in distance between consecutive simulation time instants remains bounded. Therefore, the distances obtained at simulation times are representative of the continuous-time behavior, and the distances between those times can be accurately approximated. Our derivation relies on a lifted state representation and differential inequalities. For the proposed simulation method, we provide two approaches for selecting simulation times to ensure that the error in distance approximation remains within a given bound. We then extend our method to fuel-efficiency analysis, with guaranteed error bounds for calculating the average fuel savings of vehicles. Through an example involving a highway scenario with a merging lane, we demonstrate that among string-stable control parameter settings for a vehicle platoon, some perform better in terms of safety under simultaneous packet losses and sudden braking. We also identify control parameters that result in tradeoffs between safety and average fuel savings in a vehicle platoon.

eess.SY↗
arXiv2026

iVISION-2DCD: A Long-Term Change Detection Dataset for Large-Scale Outdoor Construction Monitoring

Automation in construction is essential for reducing costs and human errors in large-scale projects. We approach the construction progress monitoring from the aspect of detecting changes in construction sites. As construction buildings continue to evolve in geometry and appearance over time, change detection need to be performed from arbitrary camera viewpoints. This necessitates developing 2D Change Detection (2DCD) algorithms that operate robustly across diverse camera perspectives at construction sites. While developing and evaluating such systems is data-intensive, no open-source benchmark dataset exists at the intersection of 2D change detection and construction automation research. Data collection using Unmanned Aerial Vehicles (UAVs) is gaining its popularity in outdoor large-scale surveying. However, in active construction sites conducting drone missions equipped with high-end sensors imposes safety concerns. Flight trajectory and collected camera viewpoints can be significantly limited. To address this critical gap, we introduce iVISION-2DCD, a large-scale synthetically generated dataset from dense LiDAR point clouds with photorealistic input images and accurate ground truth annotations. Our dataset formally defines the problem of viewpoint-robust 2DCD at construction sites and captures the inherent complexities of real-world deployment. In this paper, we present our systematic methodology for synthetic data generation, developing novel view synthesis techniques to overcome bi-temporal alignment and viewpoint diversity challenges, and implementing semi-automated semantic segmentation with change label generation while preserving challenging real-world cases. Benchmark evaluations using state-of-the-art 2DCD algorithms demonstrate that iVISION-2DCD poses novel research challenges for the computer vision and robotics communities.

cs.CV↗
arXiv2026

Cross-channel Specific Emitter Identification and Verification via Signal Envelope

Specific emitter identification (SEI) determines which known emitter a received signal originates from, while specific emitter verification (SEV) determines whether the received signal genuinely comes from its claimed emitter. In this paper, we consider the effect of wireless fading channels on SEI and SEV. When the Rician $K$-factor varies, the resulting distribution shift induced by the channel degrades both identification and verification performance. To address this issue, we first theoretically prove that the coefficient of variation of the signal envelope is strictly monotonic with respect to the Rician $K$-factor. Motivated by this property, we propose an envelope-guided adaptive feature modulation (EAFM) identifier for SEI and an EAFM with Mahalanobis distance metric learning (EAFM-MD) verifier for SEV. Specifically, the proposed EAFM identifier adopts a dual-branch neural network to extract device-oriented features from the IQ-domain input and channel-conditioning features from the normalized signal envelope, and adaptively modulates the former via feature-wise linear modulation. Then, we extend the EAFM identifier to an EAFM-MD verifier. The device-fingerprint library is constructed by storing the feature centroid and covariance for each enrolled device, along with the within-device Mahalanobis distances of training signals. For verification, the Mahalanobis distance between the extracted test features and each stored centroid is computed using the stored covariance matrix, and the minimum distance is compared to the corresponding device threshold to make a decision. Finally, numerical results show that the proposed EAFM identifier improves cross-channel identification performance, while the proposed EAFM-MD verifier achieves superior detection performance against unknown spoofing attacks.

eess.SP↗
arXiv2026

A Survey of Physical-layer Authentication Enhanced by Emerging Spatial Domain Technologies

This article surveys spatial-domain-enhanced Physical-layer Authentication (PLA), with Dual-polarized Antennas (DPA), Massive Multiple-Input Multiple-Output (MIMO), and Reconfigurable Intelligent Surfaces (RIS) as the primary focus. With the rapid growth of wireless deployments, authentication mechanisms face stringent requirements for high security, low overhead, and low latency. PLA offers lightweight identity verification by exploiting physical-layer characteristics. However, the effectiveness of PLA critically depends on how physical observations are constructed and validated under wireless channels. Unlike existing surveys that mainly organize PLA by authentication modality, feature source, and evaluation metrics, this work emphasizes the connection between spatial-domain enhancement mechanisms, the resulting feature representation, and the authentication procedure. We review how DPA, Massive MIMO, and RIS reshape PLA feature representation, and we summarize newly introduced security threats along with representative defense strategies. Case studies further illustrate the practical impact, such as representative detection-probability trends across Signal-to-Noise Ratio regimes and quantitative comparisons among representative schemes. Finally, we outline promising future opportunities enabled by Dynamic Metasurface Antennas, Extra-large MIMO, and spatial configuration with artificial intelligence.

eess.SP↗