arXiv subjects

Hao Zhu

Publications and source records attributed to Hao Zhu.

At least 19 recordsLinked to original sources

arXiv2026

ReGATE: Recurrent Gated Visual Fusion for Reactive Humanoid-Object Interaction

Humanoid loco-manipulation requires coordinating object acquisition, transport, and release from onboard sensing while responding to user commands and incomplete visual observations. We present ReGATE, a whole-body controller that combines egocentric depth, proprioception, and online locomotion and contact commands. Built on a teacher-student framework, ReGATE integrates interaction-state estimation through a recurrent adapter that predicts object geometry and physical contact without an external object-pose estimator. Head-wise and global gates regulate visual updates to a proprioceptive pathway. A visual-grounding and corruption curriculum trains the adapter, while phase-aware initialization increases coverage of brief contact transitions. In simulation, ReGATE achieves 85.2% success across 330 static-object acquisition trials, compared with 48.5% for VisualMimic under the same protocol, and 85.9% across 440 moving-object acquisition trials. Deployment on a Unitree G1 demonstrates commanded grasp-transport-release sequences and acquisition of laterally displaced boxes. A qualitative hardware experiment further demonstrates successful box acquisition when the camera is occluded after an initial observation and before the contact command. To the best of our knowledge, this is the first work to extend recurrent gated fusion of onboard depth and proprioception from perceptive locomotion to humanoid loco-manipulation under interrupted visual sensing.

cs.RO↗

arXiv2026

RoboArchitect: Automating the Design of Robotic Agents with Persistent Memory

Recent work shows that a frozen robot policy performs far better inside an agent harness, where control code decides when to query the policy, consult vision tools, or command the arm directly. Yet today these harnesses are built by hand and redesigned for each new robot or benchmark. We show that harness design can be automated, offering a strong alternative to finetuning on a novel setup. In RoboArchitect, a coding model observes a robot's rollouts, diagnoses failures, and iteratively revises the harness around the frozen policy. The harness evolves on a set of episodes for its optimization and is evaluated on held-out scenes and tasks. Without updating any model weights, RoboArchitect raises average success from 9.7% to 43.3% on unseen scenes (4.5x) and from 12.7% to 40.2% on unseen tasks (3.2x), across multiple benchmarks and policies. The gains are largest where the benchmark is far from the policy's training data: wrapping a general-purpose policy, the evolved harness yields a specialist that often outperforms even the finetuned version. Intuitive control strategies are automatically designed, such as lifting a jammed arm clear or reaching in horizontally when an obstacle blocks the way from above. The harness also evolves its own persistent memory, carrying learned experience, such as effective grasp depths, into later decisions. These results suggest that substantial headroom in robotic manipulation lies in the agent harness, and its design can be effectively automated and optimized without weight updates.

cs.RO↗

arXiv2026

RouteDRAG: Learning to Route and Control Agentic Execution for Multimodal Document RAG

Web-based applications increasingly rely on understanding long multimodal documents, where answering questions often requires combining information across pages, tables, and figures. Agentic retrieval-augmented generation addresses these tasks through iterative retrieval and reasoning. However, Agentic execution is not always necessary, introducing additional computation and potentially incorrect answers. Our empirical analysis reveals that Agentic execution provides query-dependent benefits, while additional operations may improve, preserve, or degrade answer accuracy. These findings motivate controlling both Agent invocation and subsequent execution. To address these challenges, we propose RouteDRAG, a learning-based framework comprising four components. Unified Document Index organizes multimodal content into a graph-backed index and derives query representations from initial retrieval. Adaptive Path Routing preserves exact answering and predicts the relative benefit of Agent over shorter execution paths. Dynamic Agent Execution maintains an evolving workspace and refines Planner proposals to control tool use and termination. Sequential Policy Learning first trains operation control from observed outcomes and then trains path routing using the resulting Agent's performance. Across two benchmarks, RouteDRAG achieves an average accuracy improvement of 3.47 percentage points over shorter-path-only execution while assigning only 79.3% of queries to Agent. Notably, it also surpasses uniform Agent execution on both benchmarks while requiring fewer answering calls. Our code is now available at https://github.com/Ereshkigal1/RouteDRAG.

cs.AI↗

arXiv2026

Distribution Grid Reconfiguration and Restoration via Linearizable Multi-phase Power Flow Modeling

Accurate and tractable power-flow (PF) modeling is essential for distribution grid topology optimization, where the presence of discrete branch-status decisions therein significantly increase the computational complexity. To tackle this challenge, we develop a topology-adaptive, low-complexity neural piecewise-linear (N-PWL) PF model. We propose a sequential layer design that can directly incorporate branch status variables into the PF mapping from branch flows to nodal injections. As a result, our N-PWL model not only preserves the PF branch-node consistency but also facilitates the representation of any unseen topologies without retraining. A low-complexity model design is further pursued to attain a linear NN parameter scaling by using a fixed-size area partition of the feeder. The resulting scalable N-PWL model enables mixed-integer linear programming (MILP) formulations of feeder topology optimization tasks, including distribution network reconfiguration (DNR) and restoration ordering problems (ROP). Numerical studies on the unbalanced, multi-phase IEEE 123-bus feeder demonstrate our N-PWL model outperforms existing end-to-end baselines in PF prediction and, more importantly, can fill the gap by achieving accurate representation of any unseen topologies. For solving DNR/ROP, our model can significantly improve the AC feasibility over linearized PF models, with a notable reduction of computation time compared with AC-PF based nonlinear formulations, particularly under a large number of topology decisions.

eess.SY↗

arXiv2026

Efficient SAM3 Adaptation for Multi-Class Semantic Segmentation via Dense Competitive Representations

The Segment Anything Model 3 (SAM3) has achieved significant progress in Promptable Concept Segmentation (PCS) by processing short noun phrases to generate segmentation masks with unique instance identifiers. While SAM3 excels at single-concept segmentation, it falls short in complex, real-world environments where multiple concepts must be segmented simultaneously. In particular, SAM3 typically fails to simultaneously process concepts with semantic overlap, leading to semantic misclassifications and redundant mask predictions. Moreover, as the number of concepts increases, the independent inference paradigm introduces prohibitive inference latency. To address these limitations, we propose the Dense Competitive Representations SAM (DCR-SAM) framework. By decoupling cross-modal interactions, the proposed method alleviates the inference latency. To resolve semantic misclassifications, DCR-SAM dynamically aligns textual concepts with visual representations and incorporates a dense competition head to enforce explicit inter-class competition. Furthermore, a learnable background token is applied to absorb non-target objects dynamically. These mechanisms effectively suppress overlapping predictions and generate high-fidelity masks. Extensive experimental evaluations across multiple benchmarks demonstrate that the proposed framework: (1) achieves a 2.3% $\sim$ 10.7% performance improvement over state-of-the-art methods, (2) delivers a 33$\times$ inference speedup compared to the native SAM3, and (3) exhibits strong robustness across diverse scenarios.

cs.CV↗

arXiv2026

Cooperative Boundary Switching of Mesoscopic Real Spectra in a Non-Hermitian Chain Family

At fixed bulk criticality, every fixed nonsingular real single-end potential preserves a growing local population of real eigenvalues in an open analytic family of non-Hermitian chains. An open region of two-end real potentials removes this population. A boundary-amplitude criterion fixes the minimum number of controlled ends. The population grows as $\sqrt N$, where $N$ is chain length, although its normalized spectral weight vanishes and the local limiting bands remain unchanged. For a representative chain's four zero-detuning settings, an independent passive Hermitian double chain converts this distinction into separated resonances at one fixed local port under a common coupling scan at fixed probe carrier. Its absolute reflection contrast and stability margin both scale as $N^{-1}$, with explicit sufficient resource budgets. Boundary amplitudes thus fix a minimum control requirement that limiting bands and thermodynamic spectral density do not specify.

cond-mat.mes-hall↗

arXiv2026

The noise cost of reservoir locality in broadband quantum amplifier chains

Directional quantum amplifiers enable weak-signal readout while protecting the source, but reservoir noise limits sensitivity. Finite-band suppression must reconcile frequency-dependent responses with fixed local couplings and finite rates. We construct adjacent-site loss channels attaining the known single-frequency minimum under local compatibility conditions, as an uncapped benchmark. For specified uniform and dimerized chains, we compare fixed reservoirs at matched drift, ports, high-gain band and complete gain/loss rate caps. At the representative cap $R=8$ and halfwidth $Ω=0.05$, nearest-neighbor arithmetic band average noise exceeds the unrestricted optimum by more than $1.01265$ quanta in the uniform chain for every finite $N\ge64$. Boundary-response bounds give necessary rates and targets requiring range two. At $N=64$, finite-auxiliary calculations retain a noise advantage under matched total coupling budgets; response bounds quantify noise-resolution tolerances. These theoretical results guide reservoir design and noise calibration for a proposed cryogenic microwave array with a $100$ kHz design band.

quant-ph↗

arXiv2026

TAG: Calibrating SAM3 for Universal Image Segmentation without Dense Supervision

The recent Segment Anything Model 3 (SAM3) extends powerful promptable segmentation from spatial cues to concepts, raising the prospect of a single foundation model for universal image segmentation. However, this promise often fails to translate into reliable performance on downstream segmentation tasks. In this work, we empirically identify concept recognition, rather than mask quality, as the primary bottleneck: SAM3 accurately delineates the objects it recognizes yet struggles to robustly determine which concepts are present. Driven by this, we present \textit{\underline{T}ext \underline{A}nchor re-\underline{G}rounding (\TAG{})}, a simple yet effective framework that rectifies the concept recognition of SAM3 on the target domain without dense annotations. Specifically, \TAG{} introduces an \textit{Anchor Residual Pathway} comprising image-conditioned, low-rank adapters to re-ground text anchors in the target distribution and modulate subsequent text-conditioned interactions. Guided solely by image-level category labels, \TAG{} jointly optimizes binary cross-entropy with \textit{Soft Boundary Ranking} objectives to calibrate presence confidence and distinguish present concepts from confusable negatives. Building upon this, we devise \textit{Confidence-Weighted Distillation} to couple rich semantic context of a multimodal large language model with fine-grained spatial priors of SAM3, thereby achieving effective calibration in the absence of manual annotations. Extensive experiments demonstrate that \TAG{} unlocks SAM3's segmentation potential with only 3.1M trainable parameters, achieving impressive performance across diverse scenarios, and notably, even surpassing leading mask-supervised methods on several metrics. \href{https://github.com/MiSsU-HH/TAG_SAM3}{Our code will be publicly available at GitHub \faGithub}.

cs.CV↗

arXiv2026

Distilling Graph Geometry: Knowledge Gap from GNNs to MLPs

GNN-to-MLP distillation aims to retain the predictive accuracy of a message-passing teacher while deploying a graph-free MLP at inference. Existing methods mainly transfer node-wise predictions or use confidence-based reweighting, but they do not specify where the student should preserve the teacher's graph-induced geometry. We show that this omission leads to two spectral failure modes in the student's representation space. On sparse graphs, the student suffers from spectral underfit, missing high-energy teacher directions concentrated near boundary regions. On dense graphs, it suffers from spectral overfit, retaining spurious directions that the teacher has collapsed through aggregation. Motivated by an energy-weighted teacher-student alignment objective, we propose Graph Geometry-aware MLP (G^2MLP), a training-time distillation framework guided by Ollivier-Ricci curvature. Curvature identifies where the two spectral errors concentrate and is used to allocate supervision between prediction-level and representation-level alignment. The deployed model remains a standard MLP and requires no graph access at inference. Across node-classification benchmarks, G^2MLP consistently improves over graph-free distillation baselines, reduces the teacher-student rank gap in both regimes, and transfers without architectural changes to Graph Transformer teachers and link prediction.

cs.LG↗

arXiv2026

Spectral Alignment in Forward-Backward Representations via Temporal Abstraction

Forward-backward (FB) representations provide a powerful framework for learning the successor representation (SR) in continuous spaces by enforcing a low-rank factorization. However, a fundamental spectral mismatch often exists between the high-rank transition dynamics of continuous environments and the low-rank bottleneck of the FB architecture, making accurate low-rank representation learning difficult. In this work, we analyze temporal abstraction as a mechanism to mitigate this mismatch. By characterizing the spectral properties of the transition operator, we show that temporal abstraction acts analogously to a low-pass filter that suppresses high-frequency spectral components. This suppression reduces the effective rank of the induced SR while preserving a formal bound on the resulting value function error. Empirically, we show that this alignment is a key factor for stable FB learning, particularly at high discount factors where bootstrapping becomes error-prone. Our results identify temporal abstraction as a principled mechanism for shaping the spectral structure of the underlying MDP and enabling effective long-horizon representations in continuous control.

cs.LG↗

arXiv2026

Disciplined Bilevel Programming

Bilevel optimization provides a natural modeling language for hierarchical decision problems. However, applying existing numerical solvers usually requires substantial manual analysis and reformulation. In this paper, we introduce disciplined bilevel programming (DBLP), a symbolic framework that allows users to specify and solve optimistic bilevel problems in a high-level, human-readable way that is close to the mathematical formulation. For problems with a disciplined nonlinear upper problem and a convex lower problem satisfying the disciplined parameterized programming rules, DBLP automatically canonicalizes the lower problem into conic form and constructs an equivalent single-level reformulation using the conic Karush-Kuhn-Tucker conditions. We relax the resulting complementarity constraint and use a gap continuation procedure to approximately solve a sequence of smooth nonlinear problems. We implement DBLP in the open-source Python package BLVPY, an extension of CVXPY for bilevel programming. We demonstrate the modeling and solution capabilities of BLVPY on a range of bilevel optimization problems from several application domains. The proposed framework and implementation allow users to specify and solve bilevel optimization problems within a few lines of code, without prior expertise in bilevel modeling and numerical optimization.

math.OC↗

arXiv2026

Self-Spec Verifiable Code Generation

Large language models (LLMs) may generate unreliable code on corner cases missed by testing, while formal verification can provide machine-checkable guarantees. Recently, researchers have proposed several benchmarks to evaluate the capabilities of LLMs in generating formally verifiable code, where LLMs need to formulate formal specifications, generate the corresponding code, and verify its correctness. However, existing benchmarks have two key limitations: (I) They primarily evaluate specification and code generation stage-wise, with code generation typically conditioned on an oracle specification. This setup overlooks whether strong stage-wise performance translates into end-to-end success. (II)They mainly focus on a single proof-oriented language and mathematically structured tasks, offering limited coverage of tasks common in software development. In this paper, we introduce VeriCodeBench, a benchmark for self-spec verifiable code generation, where the LLM relies solely on its own generated specification and code throughout the entire process. VeriCodeBench contains 400 language-native problems across C, Java, Rust, and Python, covering practical concerns in software development. We evaluate specification coverage, code validity, and joint problem-level success. We further introduce CodeNova to enhance the capabilities of LLMs in self-spec verifiable code generation. CodeNova makes requirements explicit through constraint-guided specification and uses verifier feedback to guide targeted implementation repairs. Experimental results reveal that self-generated specifications remain a major bottleneck, while providing more sophisticated specifications may not necessarily lead to higher verification success rates. CodeNova substantially improves performance across all evaluation metrics, enabling Claude Sonnet 5 to achieve the strongest results under the self-spec protocol.

cs.SE↗

arXiv2026

Prediction-Oriented Sequential Bayesian Design of Physical Experiments Informed by Computer Models

In many scientific and engineering domains, physical experiments are often costly, time-consuming, and limited to a certain number of design points. The Kennedy and O'Hagan (KOH) framework is widely used for combining simulator runs with limited experimental observations. Under a Bayesian implementation, simulator output, model discrepancy, and observation noise are jointly modeled by coupled Gaussian processes, followed by posterior inference and uncertainty quantification, thereby providing a probabilistic foundation for experimental design. Existing sequential design criteria mainly target better parameter estimation or global uncertainty reduction but may overlook the local structure of the predicted response. This work presents a sequential Bayesian experimental design (BED) framework aimed at improving the predictive performance of the KOH model. The framework introduces a mutual information (MI)-based criterion combined with a local complexity score that measures nonlinearity of the posterior predictive mean near each candidate point, leading to more efficient design decisions. To make the criterion computationally feasible in a fully Bayesian setting, we employ Gaussian mixture reduction with fast covariance update strategies. Through several challenging test problems and an example of the Duffing oscillator on a discrete design space, we demonstrate the effectiveness of the proposed method and illustrate its difference from conventional design baselines under sequential (offline) and adaptive (online) BED settings. The numerical results suggest that incorporating local response complexity into the MI criterion can substantially improve prediction-oriented design for physical experiments informed by computer models.

stat.CO↗

arXiv2026

Disciplined Biconvex Programming

We introduce disciplined biconvex programming (DBCP), a modeling framework for specifying and solving biconvex optimization problems. Biconvex optimization problems arise in various applications, including machine learning, signal processing, computational science, and control. Solving a biconvex optimization problem in practice usually involves heuristic methods based on alternate convex search (ACS), which iteratively optimizes over one block of variables while keeping the other fixed, so that the resulting subproblems are convex and can be efficiently solved. However, designing and implementing an ACS solver for a specific biconvex optimization problem usually requires significant effort from the user, which can be tedious and error-prone. DBCP extends the principles of disciplined convex programming to biconvex problems, allowing users to specify biconvex optimization problems in a natural way based on a small number of syntax rules. The resulting problem can then be automatically split and transformed into convex subproblems, for which a customized ACS solver is then generated and applied. DBCP allows users to quickly experiment with different biconvex problem formulations, without expertise in convex optimization. We implement DBCP in the open-source Python package dbcp, as an extension to the well known domain specific language CVXPY for convex optimization.

math.OC↗

arXiv2026

High-gain vortex transfer via activated forbidden transitions in molecular magnets

Vortex light has garnered considerable interest in recent years owing to its distinctive properties and broad application potential. In this paper, we employ a conventional three-level ladder-type configuration in molecular magnets to realize high-gain vortex light transfer by enabling otherwise forbidden transitions. We demonstrate that the intensity and phase of the generated vortex signal field are governed by the detuning of the probe field and the strength of the control field, and that the topological charges of the incident and generated fields obey a well-defined algebraic relation during the vortex transfer process. Furthermore, we show that, in molecular magnets, the Autler-Townes splitting (ATS) effect yields a larger vortex signal field gain than electromagnetically induced transparency (EIT) over a broad parameter range in nonlinear three-wave mixing. This result suggests that the widely accepted view of EIT as the dominant enhancer of nonlinear optical effects may not hold universally. In addition, we revisit previous studies and re-examine the characterization of vortex beam transfer efficiency, emphasizing the need for a more careful interpretation in multi-beam interaction systems. Leveraging the long spin coherence and microwave transitions of molecular magnets, our results may enable quantum information transfer, storage, computing, and radar imaging in solid-state platforms.

physics.optics↗

arXiv2026

Lindbladian spectral statistics beyond no-jump Hamiltonians: roles of recycling and Liouville-space structure

Spectral statistics probe integrability versus chaos and have recently been extended to Markovian open quantum systems described by Lindbladians, whose quantum-trajectory unraveling decomposes the evolution into no-jump dynamics generated by an effective non-Hermitian Hamiltonian and recycling jumps. In this work, we perform spectrum-statistics diagnostics for Lindbladians and their effective non-Hermitian Hamiltonians. We show that recycling processes, symmetry constraints, and the Liouville-space structure crucially shape the spectral correlations. In particular, we identify a family of spectrally separable Lindbladians whose spectra exhibit robust Poisson statistics, despite the effective non-Hermitian Hamiltonian varying from Poisson to strongly correlated complex-spectrum statistics. Our work establishes a unified spectral-statistics characterization for Lindbladians and their associated effective non-Hermitian Hamiltonians, deepening our understanding of spectral properties in open many-body systems.

quant-ph↗

arXiv2026

Stability of non-isentropic gaseous stars

We study the linear stability of compactly supported static spherical equilibria of the non-isentropic Euler--Poisson system. In the Schwarzschild-stable case, the linearized equations are realized as a separable Hamiltonian system on weighted spaces adapted to the physical vacuum. We prove stability against non-radial perturbations and show that the algebraic dimension of the radial unstable subspace is the Morse index of a density quadratic form subject only to the mass constraint. Thus the infinitely many linearized entropy constraints reduce, for radial perturbations, to one mass constraint and an explicit entropy reconstruction. We apply this criterion to two entropy prescriptions. No smallness condition is imposed on a fixed entropy--density relation: under the stated effective-pressure and branch hypotheses, a simple mass maximum is not a stability transition. By contrast, for a small entropy distribution fixed as a function of enclosed mass, the turning-point principle holds. In the Schwarzschild-unstable case, we construct the self-adjoint velocity operator from its quadratic form, prove a strictly negative spectral bottom and sharp exponential growth, and reconstruct energy solutions of the original first-order linearized system.

math.AP↗

arXiv2026

Generation of Custom Solvers in Rust for Convex Optimization

We introduce cvxgenrust, an open-source tool for generating custom Rust code that solves families of parameterized convex optimization problems modeled in CVXPY. cvxgenrust canonicalizes a problem family, extracts affine maps to Clarabel cone-program data, and generates a specialized Rust crate that updates parameters and calls Clarabel natively at runtime. The generated solver can also be exposed to Python and registered as a custom CVXPY solver. Our code generator supports a wide range of convex optimization problems up to semidefinite programs and exponential-cone problems. Numerical experiments show reduced runtime relative to direct CVXPY solves and performance comparable to CVXPYgen on shared problem classes.

math.OC↗