Searcharxiv⌕ Search

arXiv subjects

Wei Chen

Publications and source records attributed to Wei Chen.

At least 19 recordsLinked to original sources
arXiv2026

Integrating Machine Learning and Uncertainty Quantification for Real-Time Decision-Making in Digital Twins: A System-Centric Review and Perspective

Digital Twins are evolving from passive virtual representations into closed-loop systems that continuously synchronize with physical assets and drive real-time decisions. Yet predictive modeling, machine learning (ML), uncertainty quantification (UQ), and decision-making have been studied as separate capabilities, leaving limited understanding of their integration and maintenance during online operation. This narrative review, scoped by the functional requirements of real-time decision-making rather than by Digital Twin terminology alone, examines how such decisions are generated and sustained in engineering Digital Twins. We first examine foundational decision-making architectures and representative engineering applications, followed by methods that address their limitations across the decision loop. We identify optimization-based, policy-based, and knowledge-based approaches as the major pathways from virtual representation to action. Across them, ML plays three roles in decision-making: enabling efficient surrogate modeling, transforming heterogeneous physical feedback into decision-relevant information, and accelerating or directly generating decisions. UQ supports uncertainty-aware decisions and model adaptation, while continual validation, drift detection, and adaptation help maintain reliable operation as the physical system evolves. Together, these findings underscore the need for a shift from a model-centric to a system-centric perspective on Digital Twins, where its performance depends on the interactions with sensing, state estimation, computation, decision-making, and adaptation rather than on their standalone performance. Therefore, future research should shift from isolated algorithm development toward system-level co-design, scalable decision-making, and approaches to design, and data-efficient continual adaptation and validation for AI-enabled autonomous systems.

stat.AP↗
arXiv2026

Iterative Methods for Solving Feedback-Type Linear Equations under Small Phase Condition

This paper investigates iterative methods for solving feedback-type linear equations under small phase condition. We first develop an iterative solution scheme for linear equations with sectorial (semi-sectorial) matrices, i.e., matrices whose numerical ranges are contained in an open (closed) half complex plane. Exploiting the geometric properties of the numerical range, we introduce the notion of numerical-radius contraction as a convergence criterion. Such a scheme, with some extra effort, leads to an iterative method for the feedback-type equations under small phase condition, whose convergence condition is characterized in terms of phase information. This complements classical iterative methods for feedback-type equations based on the small gain condition. A case study on large-scale power networks is presented to illustrate the computational efficiency of the proposed method.

eess.SY↗
arXiv2026

A magnetically heated slim disc-corona model for tidal disruption events

Tidal disruption events (TDEs) can reach super-Eddington luminosities near their peak. While their optical/UV emission may arise from multiple processes, X-rays are more likely to be produced by the innermost accretion flow close to the black hole, analogous to super-Eddington flows in stellar-mass X-ray binaries (XRBs) and active galactic nuclei (AGNs). Super-Eddington AGNs typically show disc-dominated spectra, with weak and steep coronae, while super-Eddington XRBs show a wider variety of spectral shapes. Here we explore disc-corona coupling in TDEs by developing a super-Eddington ($\dot{m}>1$: slim) disc model in which magnetic reconnection transfers some of the accretion power from the disc into a hot Comptonizing corona. We test the model on the well-monitored X-ray TDE AT2021ehb, and can broadly fit the brightest X-ray spectra with a black-hole mass $M_{\rm BH}\sim10^6$--$10^7\,M_\odot$ at $\dot m\sim 1$, but need variable magnetic coupling to the disc to match the large observed fast changes in coronal contribution. AGN at similarly high luminosity do not show such variable coronae, but the transient XRB outbursts do, suggesting that it may be a common feature of rapidly changing, non-steady accretion flows. As the TDE declines to $\dot{m}\sim 0.01$, the source hardens, but never goes below $Γ\sim 2$, unlike both AGN and XRB at these luminosities. Thus the X-ray spectral evolution in this TDE does not resemble either XRB or AGN in detail, likely pointing to the very different, highly variable nature of the accretion flow in a tidal disruption event.

astro-ph.HE↗
arXiv2026

UA-Depth: Uncertainty-Aware Depth Restoration for Material-Agnostic Object Manipulation

Robotic manipulation requires accurate metric depth, yet commodity RGB-D cameras often fail on non-Lambertian objects such as transparent, reflective, and translucent materials, producing holes, structured noise, and biased depth. Most restoration methods produce only a single depth map, treating noisy and reliable regions uniformly for downstream tasks while ignoring regional uncertainty. We present UA-Depth, an uncertainty-aware depth restoration framework that iteratively co-refines depth and pixel-wise uncertainty, and uses uncertainty-gated features to suppress unreliable cues for stable updates. To strengthen zero-shot behavior, UA-Depth leverages a frozen depth foundation model as a prior to initialize missing or unreliable depth, and applies two-stage MAD-weighted log-depth scale normalization to reduce outlier bias in relative-to-metric alignment. We introduce an eight-dataset zero-shot benchmark across diverse scenes, materials, and sensors. Across these benchmarks, UA-Depth achieves the best mean rank among representative baselines, with particularly strong gains on datasets dominated by transparent and reflective objects. Beyond depth metrics, we demonstrate downstream improvements in category-level 6D pose estimation and 6D grasp synthesis by exploiting the predicted uncertainty. \emph{Code, models, and data for training and evaluation will be released upon publication.}

cs.CV↗
arXiv2026

Holistic Diversity Management in LLM-Guided Evolutionary Search

LLM-guided evolutionary search iteratively selects existing programs, uses LLMs as mutation operators to generate a new program, and evaluates it to guide subsequent search. To preserve diversity and reduce the risk of premature convergence, it often adopts the island model, in which groups of programs, called islands, evolve separately. However, this structure alone does not guarantee distinct search directions, as islands may start from similar programs or converge toward similar solutions during evolution. This suggests that diversity must be considered throughout the search process, from initializing islands with distinct programs to developing island trajectories and sharing discoveries across islands. To this end, we propose HoliDivEvolve , a holistic approach to establishing and effectively utilizing island diversity throughout evolutionary search. Specifically, we introduce diverse initialization to establish different initial programs, trajectory adaptive island selection to allocate computation based on current score, recent growth, and the remaining search budget, and growth adaptive migration to share alternative ideas across islands. Experiments across diverse benchmarks show that HoliDivEvolve achieves competitive performance on most evaluated tasks. We further conduct controlled analyses to examine how our design choices affect search behavior and final performance.

cs.CL↗
arXiv2026

Learning Nonlocal Interactions in Large-Scale Heterogeneous Metamaterials through Scientific Machine Learning

In multiscale systems, nonlocal interactions among spatially distributed constituents can unlock collective, potentially emergent behavior irreducible to individual constituent responses. Beyond characterization, nonlocality in multiscale systems is increasingly viewed as a design element across fracture mechanics, wave physics, and disordered systems. Existing nonlocality theories provide rigorous descriptions and mechanistic insight, but often require prescribed analytic forms, range, symmetries, or periodicity. Machine learning offers a data-driven route to learn interactions, yet principled approaches that tie the network design to physical theory remain nascent. Here we present domain-decomposition attention neural operator (DANO), which learns configuration-dependent interactions in heterogeneous systems governed by linear field equations from field data alone. Through a case study in large-scale optical metasurfaces, we demonstrate DANO's full-field prediction under geometric heterogeneity and across drive conditions, small-to-large domain transfer, and multifunctional inverse design, with learned interactions validated through physics-based interrogation. As an AI-for-Science tool, it supports interaction discovery and scalable, customizable design of nonlocal multiscale systems, including virtual-reality optics.

cs.CE↗
arXiv2026

ATLAS: Dual-Horizon Diagnostic Evaluation for Tool-Use Agents in Meituan Xiaotua

LLM-based assistants on web-scale service platforms increasingly address local-service needs through multi-step planning and tool use. Under dynamic business conditions and evolving user needs, outcome-level evaluation offers limited evidence of where execution fails or how service drifts across continued interaction. We present ATLAS, a dual-horizon diagnostic evaluation framework for industrial tool-use agents. Within-turn evaluation uses trajectory-wise signals across thinking and reflection, tool and skill execution, and response generation to identify execution failures and unmet capability requirements. Across-turn evaluation uses user-wise signals spanning multiple lifecycles to diagnose failures in maintaining or updating user intent, constraints, and corrections. Fine-grained signals use rule-based checks and LLM judgments with explicit evidence scopes and decision boundaries. To support reliable diagnosis at production scale, LLM judges are calibrated against dynamically maintained references from business logs, and selected judges are distilled into compact diagnostic models. The resulting diagnostic profiles help developers attribute failures, prioritize targeted interventions, and reassess improvements. Applied in Meituan Xiaotuan, a million-scale DAU local-services assistant, ATLAS achieves high diagnostic fidelity across calibrated judges and distilled models. Blinded human auditing and a 14-day online A/B test with nearly two million users per arm confirm that the ATLAS-guided agent improves recommendation relevance, factual reliability, user engagement (+2.84% answer read-through rate), and order volume (+6.13% paid orders), leading to full production deployment.

cs.AI↗
arXiv2026

One-Step Score-Based Density Ratio Estimation

Density ratio estimation (DRE) is a useful tool for quantifying discrepancies between probability distributions, but existing approaches often involve a trade-off between estimation quality and computational efficiency. Classical direct DRE methods are usually efficient at inference time, yet their performance can seriously deteriorate when the discrepancy between distributions is large. In contrast, score-based DRE methods often yield more accurate estimates in such settings, but they typically require considerable repeated function evaluations and numerical integration. We propose One-step Score-based Density Ratio Estimation (OS-DRE), a partly analytic and solver-free framework designed to combine these complementary advantages. OS-DRE decomposes the time score into spatial and temporal components, representing the latter with an analytic radial basis function (RBF) frame. This formulation converts the otherwise intractable temporal integral into a closed-form weighted sum, thereby removing the need for numerical solvers and enabling DRE with only one function evaluation. We further analyze approximation conditions for the analytic frame, and establish approximation error bounds for both finitely and infinitely smooth temporal kernels, grounding the framework in existing approximation theory. Experiments across density estimation, continual Kullback-Leibler and mutual information estimation, and near out-of-distribution detection demonstrate that OS-DRE offers a favorable balance between estimation quality and inference efficiency.

stat.ML↗
arXiv2026

SaiVLA-JEPA: Evaluating Future Latent Prediction in JEPA-Conditioned VLA Control

Joint-embedding predictive architectures (JEPAs) learn to predict representations rather than pixels, offering a compact way to model how visual scenes evolve. We ask whether this predictive objective improves closed-loop control when a vision-language-action (VLA) policy already uses V-JEPA features. We call our JEPA-conditioned flow policy family SaiVLA-JEPA. It combines a vision-language model with a diffusion transformer action expert and receives features from a frozen V-JEPA encoder. Two lightweight auxiliary heads predict future representations during training and are omitted at inference, leaving the action sampler unchanged. We compare policies without JEPA features, with current JEPA conditioning, and with additional future supervision on 10,030 paired LIBERO-Plus episodes. Current JEPA conditioning reaches about 82.5% success versus 71.6% without JEPA, while future supervision provides no detectable aggregate gain at the evaluated training budget. We separately evaluate SaiVLA-JEPA-TCA, which adapts latent context, and SaiVLA-JEPA-TPA, which adapts selected parameters. The context-retry gain is a system-level result under a larger attempt budget; parameter adaptation shows no consistent advantage over the frozen policy and target controls. These results motivate assessing latent-prediction objectives through paired closed-loop rollouts rather than latent loss alone.

cs.RO↗
arXiv2026

Optically gated transduction enables silicon electrical readout of mid-infrared light

Silicon is the dominant platform for integrated photonics, yet its wide electronic bandgap of 1.1 eV fundamentally prevents efficient detection of mid-infrared (MIR) photons through direct valence band to conduction band excitation. Existing approaches therefore seek to enhance sub-bandgap absorption or introduce external infrared absorbers. Here we demonstrate a profoundly different detection strategy based on optically gated transduction, in which MIR photons modulate a pre-established excitation state rather than directly generating photocarriers. Our device integrates Nd3+-doped lanthanide nanoparticles with an inverse-designed Al/p-Si metasurface supporting dual resonances that simultaneously enhances optical gating in the near infrared and resonantly couples incident MIR radiation. The MIR illumination redistributes the Nd3+ emission pathways, producing an optical-state change that is converted into an electrical signal through a silicon Schottky junction. The device achieves room-temperature zero-bias detection at 6.8 um with a modulation responsivity of 103 uA/W and resolves combined gate and MIR inputs into three electrically distinguishable current states. Rather than extending silicon absorption into the infrared, this work establishes gated optical transduction as a route for reading MIR optical information on a silicon platform, opening opportunities for integrated infrared sensing, signal processing and programmable photonic systems.

physics.optics↗
arXiv2026

Robustness of Anyonic Time-Domain Interference Against Dephasing

Recent anyon-collider experiments have identified time-domain braiding (TDB) as a new route to probing anyonic statistics. However, how such time-domain interference responds to environmental noise remains an open question. Here we show that TDB interference is intrinsically more robust than its spatial counterpart. In conventional spatial interference, environmental noise induces phase fluctuations that accumulate throughout all paths, leading to exponential suppression of the interference as the dephasing rate increases. In TDB, by contrast, the interfering temporal paths experience largely overlapping environmental fluctuations, whose contributions cancel out in the relative phase, so that only a small nonoverlapping temporal window contributes to dephasing. As a result, TDB interference decays only algebraically with the dephasing rate, providing a new experimental hallmark for its identification. We propose a Fabry-Perot fractional quantum Hall setup in which both types of interference coexist and experience the same environmental noise, enabling a direct comparison of their dephasing behaviors. Our results establish robustness against dephasing as a fundamental advantage of TDB and suggest new opportunities for exploiting anyonic interference in noisy mesoscopic systems.

cond-mat.mes-hall↗
arXiv2026

ClinVerse: A Realistic Multi-Agent Diagnostic Environment for Training and Evaluating Medical Agents

Disease diagnosis is a key application of large language model (LLM) agents in healthcare. However, existing medical agents are primarily trained and evaluated with static question-answering data, providing limited opportunities to learn active information acquisition, examination decision-making, and dynamic clinical interaction. To address this limitation, we present ClinVerse, an interactive, real-data-grounded multi-agent environment designed for training and evaluating medical diagnostic agents. Built on ClinPatient, a scalable dataset comprising 150k diagnostic cases derived from real-world electronic health records, ClinVerse supports coordinated interactions among six specialized clinical roles. Specifically, a doctor agent starts from an initial patient complaint, progressively gathers clinical evidence through patient interviews and examination requests, and ultimately produces a diagnosis. For each interaction trajectory, a rule-based scoring framework quantifies information acquisition, examination ordering, and diagnostic accuracy. These trajectory-level scores serve as reward signals for agent training while also enabling scalable evaluation. In experiments, we evaluate various general-purpose and medical-domain models and show that nine self-evolution methods across three categories consistently improve agents' diagnostic capabilities within ClinVerse. We further analyze the interplay among model capability, environment feedback, and self-evolution methods.

cs.AI↗
arXiv2026

CoEvolve: Counterfactual-Guided Co-Evolution of Memory and Collaboration in Multi-Agent Systems

Large language model (LLM)-based multi-agent systems have demonstrated remarkable capabilities across diverse tasks. Yet their memory and collaboration architectures often remain static, constraining continued adaptation and improvement over time. Recent work addresses this limitation by evolving memory or collaboration based on interaction experience. However, updates to these interdependent components are not necessarily aligned: an evolved workflow may require experience that memory does not provide, while evolved memory may encode guidance that the workflow cannot use. To address this problem, we propose CoEvolve, a skill-centered framework for aligned memory--collaboration evolution. CoEvolve represents reusable local collaboration patterns as skills that link evolving collaboration decisions to their supporting memories. It further compares original and counterfactual trajectories to identify local evolution signals, which coordinate updates to the skill library, skill memory, and executor memories across rounds. Experiments on the ALFWorld and WebShop benchmarks confirm that CoEvolve achieves the best overall performance among the evaluated baselines, while ablation studies further validate the effectiveness of aligned co-evolution and counterfactual-guided evolution.

cs.AI↗
arXiv2026

HiPPT: Physiology-Grounded Wearable PPG Representation Learning via Multi-Scope Text Alignment

Reusable wearable PPG representations should capture local pulse morphology and broader physiological regularities. Existing pretraining objectives, however, usually apply supervision at the recording or pair level, overlooking that different concepts become reliable over different scopes of waveform evidence. We propose HiPPT, a physiology-grounded framework that matches each supervision source to its appropriate evidence scope. We construct a PPG-Text corpus of over 4 million wrist-worn recordings from 202K users, pairing recordings with waveform-derived morphology and metadata-derived context descriptions. HiPPT combines record-level morphology alignment, which grounds individual recordings in pulse characteristics, with cross-recording context alignment, which aggregates recordings sharing the same metadata tuple to capture context-consistent regularities. Waveform-guided text modeling further recovers masked contextual concepts from PPG patch representations, providing fine-grained waveform-text supervision. Across six downstream tasks spanning blood pressure, vascular stiffness, physiological attributes, and cross-dataset heart-rate estimation, HiPPT achieves the best result on the primary linear-probing metrics. The aligned space also supports zero-shot profiling of age group, sex, and BMI category, individually and jointly, with performance improving as more recordings are aggregated. These results show that evidence-scope matching yields transferable and semantically accessible wearable PPG representations. Code and pretrained weights will be released.

cs.AI↗
arXiv2026

RubricOpt: Self-Revision of Semantic Rubrics for Preference Optimization

A model's failures can reveal gaps in the behavioral criteria guiding its training. We introduce RubricOpt, a framework that lets a policy revise these criteria within preference optimization. The policy diagnoses its own rubric-free outputs, retains requirements already covered by the rubric, and edits clauses that leave recurring failures unaddressed. A shared semantic representation connects task demands, preference-data contrasts, and rubric clauses, while a deterministic gate checks evidence references and fixed semantic anchors. The rubric then guides Direct Preference Optimization (DPO) during training, with preference pairs, labels, and the objective held fixed, and is removed at evaluation. Our analysis relates the local effect on rubric-free success to semantic alignment and a residual beyond the measured factors. Experiments on capability and open-ended generation benchmarks show that rubric conditioning can improve outcomes over ordinary DPO, with gains and regressions varying across models and datasets. A matched continuation isolates the contribution of revision from additional training and finds mixed incremental effects for the tested edit, whose feedback overlaps the evaluation set. RubricOpt thus brings the behavioral specification itself into the policy's feedback loop.

cs.AI↗
arXiv2026

Stochastic Pruning of the DPO Computation Graph: 1.4x Faster Training and Lower Peak Memory

Direct preference optimization (DPO) incurs training costs in output scoring, sequence execution, and backpropagation, so reducing supervision alone does not necessarily save computation. We introduce a unified framework for stochastic pruning of the DPO computation graph that independently samples response positions for scoring, response prefixes for execution, and preference branches for backpropagation. By making these selections before the corresponding operations, the framework omits computation at each stage of a training step. Our analysis characterizes the objective changes induced by position and sequence pruning and uses inverse-probability weighting to obtain a conditionally unbiased single-branch gradient estimator for the sampled objective. Controlled measurements show that position and sequence pruning primarily reduce peak memory, with most output-memory savings attributable to an implementation that retains full response supervision. Branch pruning yields a $1.426\times$ step-time speedup over a concatenated DPO baseline and reduces training time by about 30\% relative to stock DPO across two model families, with slightly lower mean accuracy and benchmark-dependent changes in open-ended scores. These findings relate the resource benefits of stochastic graph pruning to the computation omitted and the resulting changes in the training objective.

cs.LG↗
arXiv2026

Toward Recursive Self-Improvement in Physical AI

Recursive self-improvement (RSI) describes a system that improves not only its behavior but also the process by which it improves. Recent digital agents test this idea by rewriting their own code and checking each version with executable tests. Bringing these mechanisms into the physical world requires learning from partial observations through costly trials, while accounting for changing operating conditions and potentially irreversible consequences. We position RSI within robot learning and trace pathways from improving robot capabilities to improving the processes that generate them. We examine how physical experience can strengthen these improvement mechanisms and when the resulting gains can support further self-improvement of the mechanisms themselves. Gains carry forward only when feedback addresses remaining learning needs, the receiving component can use the change, its value survives changes in tasks or hardware, and continued practice remains safe and affordable. Existing studies, concentrated in manipulation, support behavioral improvement, benefits to later learning, and procedure improvement. Some systems also change the improver's knowledge or tools, but the physical studies reviewed here have yet to isolate gains attributable to recursion. We synthesize these findings into evaluation principles and research directions for machines that, trial by trial, safely improve the way they improve.

cs.RO↗
arXiv2026

When do machine-learning potentials need a configuration-dependent kernel? Learning long-range interactions with Fourier neural operators

Long-range interactions in machine-learning interatomic potentials are commonly treated by coupling learned sources through either a prescribed electrostatic kernel or a learned but configuration-independent one. Whether such a kernel suffices depends on the underlying screened interaction. We address this question by using Fourier neural operators (FNOs) to learn how latent source fields, built from the descriptors of a local backbone (here MACE), interact across the simulation cell. For a synthetic ionic system with tunable screening, a prescribed Coulomb kernel underestimates the long-wavelength charge fluctuations by 43% under strong screening, whereas the FNO reproduces them within 3% and recovers 95-99% of the variance of the residual left by the local backbone. A linear FNO, which learns a configuration-independent kernel, performs equally well in the homogeneous case. Under inhomogeneous screening, rearranging the remote medium modulates the coupling between distant ions, a dependence that no configuration-independent kernel can represent. Among the models tested, only the nonlinear FNO responds to this modulation, and its force errors are 27-41% lower than those of configuration-independent models. In first-principles benchmarks of liquid water, lithium lanthanum zirconate, and Au2/MgO(001), MACE-FNO is comparable to a prescribed-kernel model for the bulk systems and more accurate for the interface. It further captures the reversal of the Au2 wetting preference induced by remote Al dopants, with deviations below 5 meV per cell from density functional theory. These results indicate that a configuration-dependent kernel, as learned by FNOs, becomes necessary once the screening medium varies, even though the energy remains quadratic in the charges for a fixed medium.

cond-mat.mtrl-sci↗