arXiv subjects

He Zhang

Publications and source records attributed to He Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Topology-adaptive Estimation of Maximum Voltage Unbalance for Distribution Networks with Finite-sample Guarantee

High penetration of single-phase distributed power production and frequent topology changes can significantly increase voltage unbalance in distribution networks. Reliable probabilistic power flow (PPF) assessment for the network, therefore, must account for topology-dependent power-flow equations, numerical error in the voltage unbalance factor (VUF), and finite-sample uncertainty in power measurements. In this paper, we propose Topology-Adaptive Certified Quantile Estimation (TACQE) to estimate the maximum VUF across monitored electric buses. Following a local network change, TACQE updates the cross-couplings among power-flow sequences, and certifies candidate PPF solutions against the updated nonlinear equations. The resulted power-flow state enclosures provide rigorous bounds on the maximum VUF. Furthermore, we derive independent reference-solver evaluations to calibrate discrepancies from the reference PPF solution. This leads to a uniform sampling bound that yields simultaneous upper-tail probabilities and quantiles, which provides a finite-sample guarantee for the estimated maximum-VUF distribution over an entire range of VUF thresholds. From our numerical simulations across 101 fixed-topology quantiles, it demonstrates that the designed TACQE can provide a fast candidate post-change power flow solution where it reuses most of the previous network model and modifies only the affected bus blocks.

eess.SY↗

arXiv2026

NeuroVLA: A Brain-inspired Embodied Intelligence for Fluid and Fast Reflexive Robotics Control

Vision-language-action (VLA) models inherit powerful semantic representations from large language models, but carrying over their design priorities can leave the demands of physical control under-specified. Stable manipulation requires high-frequency proprioceptive feedback and persistent motor state, in addition to instruction understanding. Here we introduce Neuromorphic Vision-Language-Action (NeuroVLA), a brain-inspired architecture that couples a cortical vision-language planner to recurrent cerebellar modulation and a stateful spiking spinal controller. These components translate semantic intent into actions while incorporating high-frequency proprioceptive and force feedback, maintaining short-term motor memory, and enabling local responses to unexpected contact. In simulated and physical manipulation tasks, the cerebellar module reduces mean commanded jerk by 75.6%, and NeuroVLA achieves a 54.8% collision-recovery rate compared with 0.0% for the evaluated VLA baselines. The spinal controller implemented on an FPGA-based neuromorphic processor achieves 2.19 ms inference latency and 0.87 mJ per inference, corresponding to approximately 0.4 W during inference. These results motivate moving beyond inherited language-model design priorities toward embodied architectures that explicitly integrate semantic planning, high-frequency bodily feedback, motor memory, and local correction.

cs.RO↗

arXiv2026

Receiver-Constrained Recovery Placement for Encoded-Qutrit Teleportation under Monitored Decay

Using established probabilistic measurement-reversal theory, we study recovery placement in encoded-qutrit teleportation under monitored decay. We characterize the success advantage of reversing a no-jump map only on the encoded support and compare two receiver capabilities. The restricted-map construction applies to general finite-dimensional supports; explicit timing results are obtained for a coherent V-type sender, a $Λ$-type receiver, and $S=\{0,2\}$, within a prescribed family of maximal exact sender reversals. With one outcome-independent full-space receiver filter, we maximize success probability at fixed total time subject to a minimum success-weighted conditional fidelity. A single preassigned recovery time yields up to three duration-dependent regimes: no sender reversal, the earliest feasible intermediate reversal, and final-time reversal. If recovery times may be randomized independently of the input and only the pooled successful ensemble must meet the fidelity requirement, the optimum over all such timing distributions mixes the two endpoints for $ΓT \ge (\ln 3)/2$. With Bell-outcome-dependent nonunitary receiver recovery, each accepted full-column-rank transfer branch can instead be reversed exactly. For the same model and sender-operation family, omitting sender reversal uniquely maximizes the exact-teleportation success probability for every finite $T>0$ at positive decay rate. Partial pathway coherence further distinguishes relative support gain from absolute recovery success: the long-time dark-state plateau survives only for fully coherent decay. These results show how receiver control capabilities and the meaning of the fidelity constraint determine whether intermediate sender recovery is useful.

quant-ph↗

arXiv2026

Lamarck's Driving School: Discovering Autonomous Driving Training Strategies through Evolutionary Competition

Autonomous driving capabilities depend strongly on the distribution of scenarios encountered during training. Existing methods commonly construct or dynamically adapt training scenario distributions using surrogate criteria such as realism, difficulty, or risk. However, these predefined surrogates may misrepresent training value, leading to inefficient use of training resources. To address this limitation, we propose a Lamarckian evolutionary framework that replaces surrogate-based guidance with competition among candidate distributions. We formulate training strategy discovery as a multi-stage bilevel optimization problem and use Lamarckian evolution algorithm to approximate its solution. At the outer level, Darwinian crossover, mutation, and selection explore the scenario distribution space; at the inner level, policy learning acquires new capabilities, and Lamarckian inheritance transfers them to subsequent stages, allowing scenario distributions and policy capabilities to co-evolve. The resulting evolutionary trajectories reveal recurring stage-wise regularities among high-value distributions, characterized by capability accumulation through stage-wise challenge rotation. We further distill these regularities into a lightweight, reusable Lamarckian Training Strategy. Experiments show that, compared with the baseline, the complete framework reduces performance loss by up to 25.07%, while the lightweight strategy still achieves a 19.13% reduction. These results demonstrate that evolutionary competition can both discover effective training strategies and reveal reusable stage-wise patterns in how the value of training distributions changes with policy capability. Code is available on GitHub.

cs.AI↗

arXiv2026

Harnessing the Power of AI in Qualitative Research: Role Assignment, Engagement, and User Perceptions of AI-Generated Follow-Up Questions in Semi-Structured Interviews

Semi-structured interviews highly rely on the quality of follow-up questions, yet interviewers' knowledge and skills may limit their depth and potentially affect outcomes. While many studies have shown the usefulness of large language models (LLMs) for qualitative analysis, their possibility in the data collection process remains underexplored. We adopt an AI-driven "Wizard-of-Oz" setup to investigate how real-time LLM support in generating follow-up questions shapes semi-structured interviews. Through a study with 17 participants, we examine the value of LLM-generated follow-up questions, the evolving division of roles, relationships, collaborative behaviors, and responsibilities between interviewers and AI. Our findings (1) provide empirical evidence of the strengths and limitations of AI-generated follow-up questions (AGQs); (2) introduce a Human-AI collaboration framework in this interview context; and (3) propose human-centered design guidelines for AI-assisted interviewing. We position LLMs as complements, not replacements, to human judgment, and highlight pathways for integrating AI into qualitative data collection.

cs.HC↗

arXiv2026

Tri-Prompting: Video Diffusion with Unified Control over Scene, Subject, and Motion

Recent video diffusion models achieve strong visual quality and temporal coherence, but still lack coordinated control over scene layout, customized subject identity, and camera/subject motion. We study controllable video generation from three prompts: a first-frame scene image, 3D-aware multi-view subject references, and a motion-driving signal. This setting is challenging because background regions are often trackable under camera motion, while foreground subjects can rotate, self-occlude, and reveal new regions that require identity-consistent appearance. We introduce Tri-Prompting, a video diffusion framework that combines multi-view subject conditioning with dual-conditioned motion control. Tri-Prompting uses XYZ tracking points for visible background motion and low-resolution RGB proxies for foreground subject pose, allowing the model to recover fine appearance from multi-view references while retaining flexibility for plausible subject-scene interactions. An inference-time ControlNet scale schedule further balances motion controllability and visual realism. We evaluate Tri-Prompting against specialized baselines: DaS for motion reconstruction and Phantom for subject-driven generation. Tri-Prompting achieves competitive or improved reconstruction quality, stronger multi-view identity preservation, and better 3D consistency, while enabling 3D-aware subject insertion and in-scene manipulation.

cs.CV↗

arXiv2026

Traverse: Learning When to Remember, Reset, and Redirect for Long-Horizon Web Search

Long-horizon information-seeking agents often accumulate noisy or misleading context, causing early mistakes to persist and making recovery increasingly difficult. We introduce an autonomous search harness in which the agent manages its own search process through three states: Rubric, Answer, and Verify. The agent first defines criteria for a valid answer, searches under these criteria, and then independently verifies the result before deciding whether to terminate or continue searching. It is further equipped with a Seal Memory tool that enables active context management. Training this behavior with reinforcement learning, however, can induce Seal Collapse, resulting in unstable training and preventing the agent from reliably learning when and how to use its memory tools. We solve this with a simple strategy that trains only the final segment after context management. Our 35B model achieves 72.83 on BrowseComp, outperforming comparable open-source systems, and consistently improves over the base model across BrowseComp-ZH, xbench, DeepSearchQA, WideSearch, financial investigation, and product search. Ablations show that autonomous compression outperforms automatic compaction and validate our RL design.

cs.CL↗

arXiv2026

COMPASS: Predicting the Relationship of Multiple Patches for Vulnerabilities with LLMs

Modern software heavily relies on code reuse, so upstream vulnerability fixes do not automatically propagate to downstream codebases. Downstream maintainers must manually adopt patches to eliminate known risks. In practice, a single vulnerability often corresponds to multiple patches, which greatly complicates downstream patch adoption because different patch relationships imply different adoption strategies. To address this challenge, we first manually inspect large-scale multi-patch vulnerabilities (about 1K) in the real world and interview experienced developers, summarizing six typical types of patch relationships, i.e., Merge, Mirror, Better Solution, Fixing-of-Fixing, Collaboration, and Separation. Based on these observations, we propose COMPASS, an automated approach that predicts the relationships of multiple vulnerability patches with large language models. Given a CVE as input, COMPASS follows a four-phase pipeline that (i) identifies the patch group and pre-scans explicit relationships, (ii) performs individual patch analysis, (iii) infers relationship instances via a hierarchy-guided prompt, and (iv) validates completeness and consistency of the inferred results. As output, COMPASS reports the predicted relationships within the patch group and visualizes them as a relationship graph. We evaluate COMPASS on a benchmark of 300 multi-patch CVEs and compare it against mainstream learning-based and LLM baselines. Results show that our method achieves strong and consistent prediction effectiveness and outperforms SOTA by 85.04% on average. We publicly release an online querying website to support community reuse of patch relationships knowledge: https://patch-relation.com.

cs.SE↗

arXiv2026

Source-Lifted Flow Matching for Intervenable Multimodal Imitation

Flow-matching policies are promising for imitation learning because they model complex multimodal action distributions. However, their stochasticity is largely passive: repeated sampling may yield diverse behaviors, but users cannot directly choose among valid continuations from the same state. We propose Source-Lifted Flow Matching (SL-FM), a source-intervenable flow-matching policy that exposes such a handle while keeping the velocity field shared and latent-free (without a separate discrete-handle input). The handle selects only the source endpoint of the conditional flow, not a mode-specific field, preserving the standard formulation while avoiding decomposition into separate mode-conditioned dynamics. The core mechanism is Orthogonal Source Lifting, designed to prevent path-crossing ambiguity. Instead of partitioning target actions by mode, SL-FM lifts handle-specific sources into auxiliary orthogonal coordinates and keeps targets in the original action subspace. This preserves the demonstrated action distribution while allowing one shared field to carry different branches without merging at crossings. To keep handles usable across states, we learn a state-dependent source mixture end to end and use a responsibility floor, giving each handle weak supervision and mitigating dead modes. Experiments on crossing-flow diagnostics and robot-control benchmarks show that SL-FM converts passive source randomness into an actionable intervention variable. It removes crossing-induced composite trajectories, changes future routes in 91.1% of matched-prefix interventions, and achieves strong free-deployment performance, with improvements in several benchmark settings. Overall, source geometry provides actionable multimodal control without conditioning the velocity field on the selected mode.

cs.RO↗

arXiv2026

Universal Pose Pretraining for Generalizable Vision-Language-Action Policies

Existing Vision-Language-Action (VLA) models often suffer from feature collapse and low training efficiency because they entangle high-level perception with sparse, embodiment-specific action supervision. Since these models typically rely on VLM backbones optimized for Visual Question Answering (VQA), they excel at semantic identification but often overlook subtle 3D state variations that dictate distinct action patterns. To resolve these misalignments, we propose Pose-VLA, a decoupled paradigm that separates VLA training into a pre-training phase for extracting universal 3D spatial priors in a unified camera-centric space, and a post-training phase for efficient embodiment alignment within robot-specific action space. By introducing discrete pose tokens as a universal representation, Pose-VLA seamlessly integrates spatial grounding from diverse 3D datasets with geometry-level trajectories from robotic demonstrations. Our framework follows a two-stage pre-training pipeline, establishing fundamental spatial grounding via poses followed by motion alignment through trajectory supervision. Extensive evaluations demonstrate that Pose-VLA achieves state-of-the-art results on RoboTwin 2.0 with a 79.5% average success rate and competitive performance on LIBERO at 96.0%. Real-world experiments further showcase robust generalization across diverse objects using only 100 demonstrations per task, validating the efficiency of our pre-training paradigm.

cs.CV↗

arXiv2026

A divergence-free parametric finite element method for 3D Stokes equations on curved domains

The Stokes equations play an important role in the incompressible flow simulation. In this paper, a novel divergence-free parametric mixed finite element method is proposed for solving three-dimensional Stokes equations on domains with piecewise smooth boundaries. The flow velocity and pressure are discretized with high-order parametric Brezzi-Douglas-Marini elements and volume elements, respectively, on curved tetrahedral meshes. Utilizing the interior-penalty discontinuous Galerkin (IPDG) technique, we prove the inf-sup condition for the mixed finite element pair, and high-order optimal error estimates in the energy norm, with the help of the extension and transformation of the true solution to computational domain. Moreover, the discrete velocity is exactly divergence-free, meaning that $\Div\Bu_h=0$ holds in the curved computational domain. Numerical experiments are conducted to support the theoretical analyses.

math.NA↗

arXiv2026

You Don't Need To Train: Agentic Heuristic Learning Studio for Executable Human Activity Recognition

Human activity recognition (HAR) is usually framed as gradient-based training of neural networks. Agentic Heuristic Learning (AHL) Studio explores a complementary view inspired by human cognitive learning: people learn activities by remembering examples, forming rules, and repairing mistakes, not by backpropagating. This proposed tool implements AHL for HAR: a learning-time agent reasons over sensor protocols, proposes executable heuristic policies, records repair traces, and exports an LLM-free policy for edge deployment. We focus on the HAR benchmark family and provide an end-to-end workflow from dataset observation to edge-oriented export. On eleven HAR datasets evaluated so far, AHL policies reach strong executable-policy performance while remaining inspectable, editable, and replayable \footnote{https://github.com/zhaxidele/ahl-ts-studio}.

cs.LG↗

arXiv2026

EdgeHAR: An Edge-Native Compact Sensor Foundation Model for Human Activity Recognition

Sensor-based human activity recognition (HAR) is fundamental to ubiquitous and wearable computing, yet existing foundation models are largely designed for cloud-scale deployment and struggle with real-world sensing shifts, including unseen users, devices, sampling rates, and sensor placements. We present \textbf{EdgeHAR}, an edge-native compact sensor foundation model designed for wearable intelligence. Unlike conventional models that entangle activity knowledge with acquisition variations, EdgeHAR learns transferable representations by factorizing sensor signals into three latent codes: an \textbf{(i)Activity-Semantic Code} capturing reusable activity knowledge, a \textbf{(ii)Motion-Dynamics Code} modeling temporal patterns, and an \textbf{(iii)Acquisition-Context Code} representing sensor-specific variations. This disentangled design enables efficient adaptation to new users, devices, placements, and activity classes with limited target-domain data. By incorporating lightweight adaptation modules, EdgeHAR achieves foundation-model-level transferability while satisfying edge constraints in computation, memory, latency, and privacy. Experiments across heterogeneous HAR datasets demonstrate that EdgeHAR maintains competitive recognition performance under distribution shifts with substantially reduced deployment cost. EdgeHAR establishes a practical paradigm for compact, edge-first foundation models for ubiquitous sensing systems.

cs.LG↗

arXiv2026

Kairos: A Dataset for Fine-Grained Video-Language Modeling over Space, Time, and Dynamics

Many emerging video language modeling tasks require systems to move beyond clip-level abstraction and model visual content as it unfolds over extended time horizons. However, most existing video datasets rely on coarse or sparsely aligned supervision, which compresses temporal variation and limits the ability of models to learn reusable representations of continuous visual dynamics. We introduce Kairos, a video dataset for video-language modeling with time-resolved annotations. Kairos consists of long-duration videos, ranging from ten minutes to half an hour, annotated with fine-grained temporal alignment. The annotations capture ongoing actions, entity appearances and attributes, interactions, and evolving contextual cues along the video timeline. This time-resolved structure supports fine-grained evaluation, long-range modeling and reasoning, instruction data construction, representation learning, and video generation. Kairos provides a general-purpose foundation for modeling visual experiences over time.

cs.CV↗

arXiv2026

Inversion Framework of Internal Mass Distribution Parameters of Asteroid Apophis from Dynamical Observations

The detection of the internal mass distribution of asteroids is of great significance for understanding their origin, evolution, and mission planning for exploration. Previous approaches rely on indirect density estimates or close-range spacecraft gravity inversion, which have limited applicability. This paper presents a proof-of-concept framework to infer the internal mass properties of asteroid (99942) Apophis during its close Earth flyby in 2029 using dynamical observations collected during the encounter. We establish a dynamical mapping from the evolution of orbital and rotational states to internal structural parameters, formulate it as an inverse problem, and solve it using Particle Swarm Optimization. The algorithm is first validated on a regular ellipsoidal model and then applied to three mass distribution models based on the actual shape of Apophis. Under ideal observation conditions, the relative error of the inverted moment of inertia ratios can be below 0.001%, and the absolute error of the center-of-mass position reaches the order of 10-5 meters. The algorithm successfully distinguishes among different internal structures. When realistic measurement noise is introduced, the inversion accuracy degrades. A sensitivity analysis reveals that the accuracy of the inertia tensor inversion is primarily limited by angular velocity measurement noise, whereas center-of-mass determination is highly sensitive to the precision of position and velocity data. This study provides a proof-of-concept for a technically feasible and cost-effective approach to infer asteroid internal structure during close encounters, and also highlights the critical data accuracy requirements for practical application, offering guidance for future observation campaigns.

astro-ph.EP↗

arXiv2026

When Should a Network Emit Geometry, and When Should It Detect It? Readout, Reconciliation, and Representation in Floorplan Vectorization

A network trained to recover the walls, openings, and rooms of a rasterized floorplan can produce its output in two ways: by emitting the geometry as an autoregressive coordinate sequence, or by detecting it on dense junction and centerline heatmaps and assembling a graph. We compare the two readouts on the same trained network. On real scans (CubiCasa5K) detection is better on every wall measure (+2.7 wall F1 at tolerance 0.05, +5.1 at 0.015; paired bootstrap intervals exclude zero), and reading an opening heatmap the decoder never used raises opening F1 by 2.6x without retraining. Within real scans the readout's advantage grows with plan size and reverses on small plans; on clean vector renders sequence decoding is better by 5 to 8 points where its training covered the render style, while under full domain shift the readout, given calibrated thresholds, stays ahead; neither ink density nor plan size explains the reversal. With matched data and recipe, a room-centric system with a reconciliation step and a wall-first sequence model reach comparable wall quality, so the output representation matters less than is usually assumed. A prior from the other family helps at the output but not at the input: deterministic fusion of the two outputs raises wall F1 by 7 points, whereas conditioning one model on the other's output gives no gain in three forms, including two ground-truth-content controls. We also provide an edit-cost metric that scores a draft by the human work needed to correct it, corrected CubiCasa5K annotations, and ResPlan-FP, a CC BY 4.0 benchmark of 16,998 plans with frozen splits and three baseline tracks. Code, the benchmark, and the corrected annotations are available at https://github.com/Cyprinus12138/fpvec-lab

cs.CV↗

arXiv2026

Constructing edge-disjoint Steiner trees in Cartesian product networks

Cartesian product networks are always regarded as a tool for ``combining'' two given networks with established properties to obtain a new one that inherits properties from both. For a graph $F=(V,E)$ and a set $S\subseteq V(F)$ of at least two vertices, \emph{an $S$-Steiner tree} or \emph{a Steiner tree connecting $S$} (or simply, \emph{an $S$-tree}) is a subgraph $T=(V',E')$ of $F$ that is a tree with $S\subseteq V'$. For $S\subseteq V(F)$ and $|S|\geq 2$, the {\it generalized local edge-connectivity} $λ(S)$ is the maximum number of edge-disjoint Steiner trees connecting $S$ in $F$. For an integer $k$ with $2\leq k\leq n$, the {\it generalized $k$-edge-connectivity} $λ_k(F)$ of a graph $F$ is defined as $λ_k(F)=\min\{λ(S)\,|\,S\subseteq V(F) \ and \ |S|=k\}$.In this paper, we give sharp upper and lower bounds for $λ_k(G\Box H)$, where $\Box$ is the Cartesian product operation, and $G,H$ are two graphs.

math.CO↗

arXiv2026

Tracing Provenance and Detecting Tampering with Complementary LLM Watermarks

Watermarking LLM-generated text is an important task for tracing its provenance. Existing LLM watermarks preserve provenance under editing, but this same robustness allows an adversary to alter critical content while retaining attribution, a vulnerability known as piggyback spoofing. We introduce an innovative watermark that jointly provides provenance and tamper evidence. It co-embeds a robust signal and a fragile signal into each generated token. The signals share the same mechanism but use independent keys and different seeding windows over normalized text, making one resilient to edits and the other sensitive to reader-visible changes. Multiple rounds of unbiased tournament reweighting preserve the expected generation distribution, while a periodic round-allocation pattern controls the trade-off between the two signals. At detection, their scores form a two-dimensional space supporting three decisions: Intact, Tampered, and No-Watermark. Across two large language models and two prompt datasets, our method demonstrates the highest tamper-detection rate among the evaluated methods while maintaining competitive attribution robustness and perplexity. Ablation studies show that reliable three-state detection requires a well-defined notion of intactness, co-embedding of the two signals, and complementary sensitivity to edits.

cs.CR↗