Searcharxiv⌕ Search

arXiv subjects

Hao Zhu

Publications and source records attributed to Hao Zhu.

At least 19 recordsLinked to original sources
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↗
arXiv2026

The PIMMUR Principles: Ensuring Validity in Collective Behavior of LLM Societies

Large language models (LLMs) are increasingly used to simulate human collective behavior, yet claims that such simulations are human-like remain largely untested. We conducted a systematic audit (pre-registered on OSF) of LLM-based social simulations across four databases (Scopus, IEEE Xplore, ACM Digital Library, and arXiv). Across 576 studies reported in 350 recent papers, we applied six methodological evaluations: agent Profile, Interaction, Memory, Minimal-Control, Unawareness, and Realism (PIMMUR). Coding every study against pre-specified rules, we revealed that PIM were met more often than MUR. Frontier LLMs correctly identified the underlying social experiment in 65.2% of cases, and 50.6% of prompts imposed constraints that pre-determined the outcome. These compliance rates are upper bounds, because incomplete methodological reporting (for example, unreleased prompts) limits the available evidence. Reproducing five representative experiments (e.g., opinion dynamics), we found that reported collective phenomena often vanish or reverse once PIMMUR principles are enforced, indicating that many "emergent" behaviors are methodological artifacts rather than genuine social dynamics. Current LLM simulations may therefore capture model-specific biases rather than universal features of human social behavior, raising concerns about their use as scientific proxies for human society.

cs.CL↗
arXiv2026

ASTRA: Asynchronous Spatio-Temporal Reconstruction via Trajectory Alignment

Dynamic 3D scene reconstruction has made significant progress with multi-camera systems, often relying on temporally aligned observations across views. However, in real-world scenarios, temporal asynchrony among capturing devices remains a common limitation, leading to severe motion blur and geometric artifacts. Existing asynchronous reconstruction methods typically estimate temporal offsets through photometric supervision, but appearance matching provides weak temporal cues under large offsets and complex motions. We attribute this limitation to two critical issues: texture-induced collapse, where low-textured regions provide nearly vanishing alignment signals, and deformation-induced entanglement, where temporal errors are absorbed into distorted geometry or motion rather than being explicitly corrected. To address these issues, we propose ASTRA (Asynchronous Spatio-Temporal Reconstruction via Trajectory Alignment), a framework that introduces 2D motion trajectories as explicit, texture-robust supervision for asynchronous dynamic reconstruction. Instead of synchronizing cameras solely through rendered color residuals, ASTRA jointly optimizes temporal offsets and dynamic 3D representations by aligning the projected motion of reconstructed 3D points with observed 2D trajectories, while using dynamic and certainty masking to suppress unreliable trajectory constraints. Extensive experiments on different dynamic Gaussian Splatting backbones show that ASTRA preserves high-frequency spatial details and sustains strong robustness even under severe asynchrony with up to 25-frame offsets, achieving approximately 1.4 dB PSNR improvement, reducing temporal-offset MAE by 54.0%, and nearly quadrupling the synchronization success rate.

cs.CV↗
arXiv2026

RouteGraph-Mona: Confusion-Aware Routing Fine-Tuning for Mineral Image Classification

Mineral image classification is important for geological exploration and resource development, but it remains challenging due to substantial intra-class variations in appearance and high inter-class visual similarity. Multi-cognitive Visual Adapter (Mona) is a vision-oriented parameter-efficient adapter that adapts pre-trained visual models by tuning only a few parameters. However, Mona statically aggregates responses from multiple scales, limiting its ability to accommodate sample-specific scale preferences and model confusion among visually similar mineral categories. To address this issue, we propose \textbf{RouteGraph-Mona}, a lightweight route-space regularization method built on Mona. Specifically, we replace Mona's static multi-scale aggregation with sample-adaptive routing. The resulting branch-selection behavior defines a compact routing space that captures each image's scale preferences. We then regularize the resulting routing signatures with class-wise route anchors and confusion-weighted margins. The route anchors encourage class-consistent routing patterns, while the margins promote greater separation between visually similar categories in the routing space. Experiments on three public mineral image datasets with two visual backbones show that RouteGraph-Mona consistently outperforms Mona in mean accuracy and remains competitive with representative fine-tuning methods and mineral image classification baselines.

cs.CV↗
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

Critical Weather Scenario Screening Using Weather-to-Voltage (W2V) Predictive Modeling

This paper proposes a critical weather scenario screening framework for identifying weather conditions that can trigger high-voltage (HV) events in the power grid. Unlike conventional weather-aware contingency analysis limited to component-level outage risk, our framework screens weather scenarios as potential drivers of grid-level voltage violations. Given a non-critical weather scenario, we seek the perturbation over the high-dimensional weather space to maximize a pre-defined voltage criticality score, by using a differentiable weather-to-voltage (W2V) predictive model to facilitate the gradient update over a compact latent space. Specifically, a non-negativity constraint is used for achieving physically-consistent perturbations, with another L1-norm based constraint for bounded perturbation. The latter could promote sparse and interpretable perturbations, and this uniform budget also yields a sensitivity-aware vulnerability ranking across different weather scenarios. Numerical experiments on a 6717-bus synthetic Texas system have effectively demonstrated the potential of weather uncertainty in triggering HV events, and this potential cannot be represented by the voltage analysis of individual weather scenarios. Interestingly, the most vulnerable scenarios are characterized by wind-dominated perturbation patterns concentrated in high wind-capacity regions, coinciding with observations from actual power flow data and experiences in real system operations.

eess.SY↗
arXiv2026

DiGS-Avatar: Single-Image Animatable 3D Human Reconstruction via UV-Space Diffusion

Single-image 3D human reconstruction often suffers from over-smoothed textures and geometric inconsistencies. While diffusion models improve generative quality, their reliance on multi-view synthesis prior to 3D reconstruction is computationally expensive and prone to view inconsistency. We propose DiGS-Avatar, which reformulates this task as an efficient, diffusion-based UV-latent completion task, ensuring 3D consistency by design. To capture accurate spatial structure, we introduce a teacher-student framework where a multi-view teacher provides geometrically aligned pseudo-ground-truth latents to supervise a single-view diffusion student. Treating this inferred latent as a robust structural skeleton, our method injects high-level semantic features to accurately recover fine textural details without disrupting spatial integrity. The refined representation is then decoded into 3D Gaussian primitives. Extensive experiments demonstrate that DiGS-Avatar achieves state-of-the-art or highly competitive visual fidelity and zero-shot generalization, while reconstructing a fully animatable 3D avatar in just 0.71 seconds. Code is available at https://github.com/KLMAV-CUC/DiGS-Avatar.

cs.CV↗
arXiv2026

PRAXIS: Graph-Grounded Tacit Knowledge for Domain Code Generation

LLM agents have achieved strong performance on general software engineering tasks, yet struggle with domain-specific code generation. We identify the root cause as the agent's lack of tacit knowledge, including domain-specific business rules, interface contracts, and operational conventions that developers internalize through practice but never document. This knowledge is deeply buried beneath the domain code, dispersed across code entities and their dependency relations, and invisible to the agent that lacks it. These properties make tacit knowledge inherently difficult to retrieve or learn. In this work, we propose PRAXIS, a framework that enables agents to systematically extract, represent, and reuse tacit knowledge for domain code generation. PRAXIS acquires tacit knowledge by simulating human development workflows within the target codebase, distills it into structured units organized on the code dependency graph, and proactively surfaces it to the agent at the point of code interaction. Extensive experiments demonstrate that PRAXIS outperforms state-of-the-art agents equipped with powerful agentic search capabilities, as well as experience-based and skill-based methods. The approach integrates seamlessly into various agent frameworks and LLMs with consistent performance improvements, and supports continual evolution with performance steadily scaling as practice accumulates.

cs.SE↗
arXiv2026

ReX-Shot: Single-Image Rephotography via Geometry- and Camera-Grounded Generation

Single-image rephotography aims to synthesize new shots of a scene from a single reference image with specified viewpoints, focal lengths, and photographic effects, which are intrinsically coupled in imaging. Existing methods typically treat these factors separately and struggle under joint control: novel-view synthesis may introduce geometric distortions under focal-length changes, while super-resolution and instruction-guided editing remain confined to 2D and cannot reliably extend detail restoration or appearance control to novel viewpoints. We attribute these limitations to imperfect single-image 3D reconstruction and the sampling limit of continuous focal-length enlargement. To reduce projection bias from geometric errors, we use implicitly transformed foundation-model features for robust target-view guidance. We further formulate focal-length enlargement as a geometry-guided super-resolution problem and exploit generative detail priors to recover details lost during sparse 3D resampling. Built on this 3D-aware generative backbone, we lift photographic-effect control from 2D filtering to 3D-aware appearance editing, preserving content consistency across viewpoints and focal lengths. These components form ReX-Shot, a geometry- and camera-grounded generative framework for single-image rephotography. To our knowledge, ReX-Shot is the first unified framework to jointly control viewpoint, focal length, and parameterized photographic effects from a single image. Experiments show that ReX-Shot outperforms representative baselines across all three controls while enabling near-real-time interactive rephotography.

cs.CV↗
arXiv2026

One-dimensional Polar Spinor Droplets

We derive a channel-resolved Lee-Huang-Yang correction and construct an extended GrossPitaevskii model for one-dimensional polar spin-1 quantum droplets. The fluctuation contribution separates into density and spin channels, which supports self-bound droplets even when the spinindependent mean-field interaction is repulsive. Stationary solutions exhibit a continuous crossover from soliton-like to flat-top droplets, accompanied by saturation of the chemical potential and peak density as the particle number increases. Within the parameter range examined here, linear Bogoliubov analysis together with weak-perturbation dynamics supports the stability of both droplet types. A quadratic-Zeeman quench reveals a finite-size crossover in breathing dynamics and distinct nonequilibrium roles of the density and spin fluctuation channels. Representative head-on collisions further show that the finite-size crossover modulates phase-sensitive nonlinear scattering, with in-phase impact producing coalescence-like central retention and out-of-phase impact favoring quasi-elastic separation. The analysis clarifies how density and spin fluctuations shape equilibrium structure and nonequilibrium response in low-dimensional polar spinor droplets.

cond-mat.quant-gas↗
arXiv2026

StartupBench: Benchmarking General-Purpose Agents on Market-Validated End-to-End Workflows

Recent advances in Large Language Models(LLMs) and agents have substantially improved the ability of AI systems to execute complex tasks. Yet existing benchmarks largely rely on researcher-selected tasks, leaving uncertain whether such progress extends to the work that real-world users actually demand from AI systems. We introduce \textbf{StartupBench}, an E2E agent benchmark grounded in market-validated AI startup products. Rather than defining tasks from pre-defined assumptions about useful agent capabilities, we systematically study AI products with demonstrated adoption, together with their product workflows and users, to identify real-world tasks for which AI has established practical demand across diverse professional domains. We translate these workflows into complete deliverable-oriented tasks and evaluate them with fine-grained rubrics capturing their complex requirements. Across representative models evaluated under a unified agent harness, even the strongest model successfully completes only approximately 30\% of StartupBench, despite making substantial partial progress on many tasks. Further analysis identifies aspects like complex instruction following and domain-specific expertise as major sources of failure. Our results reveal that many market-validated workflows remain beyond the reliable capabilities of current general-purpose agents, establishing StartupBench as an empirical measure of progress toward E2E completions of real-world user tasks.

cs.AI↗
arXiv2026

GraphQLer: Enhancing GraphQL Security with Context-Aware API Testing

GraphQL APIs power production systems across financial services, e-commerce, and social platforms, yet their most critical access-control vulnerabilities--Insecure Direct Object Reference (IDOR), Use-After-Free (UAF), and state-dependent injection--routinely escape automated security testing. Industry-standard scanners (ZAP) and the leading research fuzzer (EvoMaster) test operations in isolation and cannot compose the multi-step sequences these flaws require. We present GraphQLer, an open-source automated security testing framework built for production GraphQL APIs. GraphQLer constructs a typed dependency graph from live schema introspection and synthesizes vulnerability chains--ordered operation sequences targeting specific flaw classes. Three strategies cover the critical attack surface: topological SCC-traversal for general reachability, cross-user IDOR replay for broken access control, and CREATE -> DELETE -> READ synthesis for UAF. On a production financial API (FinServ), GraphQLer identified eight potential vulnerabilities--including denial-of-service vectors that exposed stack traces and sensitive implementation details--without prior documentation or authentication credentials. On a self-hosted Saleor instance pinned to the CVE-2022-39275 commit, GraphQLer reproduced all four broken-access-control mutations cited in the security advisory. On the 11 public APIs of the coverage set, GraphQLer achieves 85.52% mean PositiveCoverage versus 29.29% (EvoMaster) and 21.80% (ZAP); across the 21 evaluated APIs it detects all 5 confirmed IDOR endpoints, UAF behavior on two controlled schemas, and confirms XSS and SQLi on DVGA (an independent third-party oracle)--while all baselines detect zero chain-based vulnerabilities.

cs.CR↗
arXiv2026

Generation of vortex-squeezed light in a coherently prepared medium

In this Letter, we theoretically propose an alternative scheme for generating vortex optical squeezing, based on the Raman scattering process in a coherently prepared medium, distinct from approaches such as parametric down-conversion and four-wave mixing. Our analysis reveals that both the input control field and the generated signal field can exhibit squeezing upon adjusting the relevant system parameters, with the control field exhibiting a greater degree of squeezing under identical conditions. We further demonstrate the existence of optimal squeezing values over a range of tuning parameters, highlighting the flexibility and robustness of the proposed scheme. These findings may offer a new reference for continuous-variable vortex optical squeezing and possess potential applications in domains such as quantum information processing, quantum precision measurement, and quantum sensing.

quant-ph↗
arXiv2026

Propagation dynamics of high-gain vortex beams in symmetry-broken media via forward and backward three-wave mixing

In recent years, vortex light, a distinctive form of structured light, has attracted considerable attention owing to its unique properties and the rich physical phenomena arising from its interaction with matter. In this paper, we investigate the propagation dynamics of vortex beams in a symmetry-broken three-level system based on forward and backward three-wave mixing (TWM) processes. We find that both processes enable the transfer of high-gain vortex light, with the associated topological charges obeying identical algebraic relations. The forward process exhibits periodic oscillatory transmission and modulates the transverse spatial profile of the generated signal vortex field, whereas the backward process features stable transmission and produces a signal field with higher gain and improved fidelity. Under the Autler-Townes splitting (ATS) regime, the probe field detuning in both schemes has a negligible influence on the gain. Optical depth influences only the rate at which the gain approaches its maximum, while the peak value remains unchanged. These results constitute a meaningful extension of the work reported in reference [40] and may provide a feasible approach for quantum communication, quantum computation, and the generation of high-gain, high-fidelity vortex light.

quant-ph↗
arXiv2026

FastTPS: An Optimized Method for LLM Token Phase for AI accelerators

The popularity of large language models (LLMs) escalates an ongoing demand for effective inference. However, due to the sequential processing of tokens during the token phase in decoder-only LLMs inference, the inherent low parallelism leads to reduced throughput and suboptimal utilization of the computing units on artificial intelligence (AI) accelerators, particularly when handling long-sequence inputs that impose significant memory overhead. Recently, many reported methods have been developed as potential solutions, since they emerge with numeric deviation. This paper presents FastTPS, a high performance and low-precision loss method for accelerating the token-phase in LLM inference on general AI accelerators which includes three key components: (1) AI accelerator-enabled reloading-free KV Cache concatenation which decreases memory access overhead as well as enables full fusion of Attention, (2) high-efficiency and high-accuracy 'RoPE' attention based on the tiling optimized FLAT, and (3) highly-fused MLP with fine-grain pipeline scheduling. Our results confirm that FastTPS significantly alleviates memory bottlenecks in the token phase, delivering a 6x speed improvement (compared to none-fusion) on an AMD Ryzen AI 300 series NPU with BF16 precision while sustaining 93% peak memory bandwidth utilization during Phi3-mini-4k-instruct inference.

cs.LG↗
arXiv2026

AutoMem: Automated Learning of Memory as a Cognitive Skill

Memory expertise is a learned skill: knowing what to encode, when to retrieve, and how to organize knowledge--a capacity known in cognitive science as metamemory. We bring this perspective to LLMs by treating memory management as a trainable skill. We promote file-system operations to first-class memory actions alongside task actions, letting the model itself decide how to manage its memory. This memory skill improves along two axes: the structure that supports it (prompts, file schemas, action vocabulary), and the proficiency of the model exercising it. Both axes resist manual optimization: episodes in long-horizon tasks run for thousands of steps, and a single memory mistake can hide long before it surfaces, making human review of full trajectories impractical. We introduce AutoMem, a framework that automates both axes. In the first loop, a strong LLM reviews complete agent trajectories and iteratively revises the memory structure that shapes how the agent interacts with its memory files. In the second loop, the agent's own good memory decisions are identified from many episodes and used as training signal to sharpen the model's memory proficiency directly. Across three procedurally generated long-horizon games (Crafter, MiniHack, and NetHack), optimizing memory alone--without modifying the model's task-action behavior--improved the base agent's performance ~2x-4x, bringing a 32B open-weight model competitive with frontier systems such as Claude Opus 4.5 and Gemini 3.1 Pro Thinking. Our results show that memory management is an independently learnable skill, and a high-leverage objective yielding large gains on long-horizon tasks.

cs.AI↗