Searcharxiv⌕ Search

arXiv subjects

Gang Wang

Publications and source records attributed to Gang Wang.

At least 19 recordsLinked to original sources
arXiv2026

Reveree: Diagnosing LLM Reverse-Engineering Agents

Reverse engineering (RE) is critical to security tasks such as malware analysis and vulnerability discovery, and large language model (LLM) agents are increasingly able to perform it autonomously. Capture-the-flag (CTF) RE challenges have become the standard proxy for measuring this capability, but evaluation rests on a single criterion: whether the agent captures the flag. This solve rate reveals neither where in the RE process an agent fails nor whether a success reflects analysis of the binary or recall of a public solution. In this paper, we propose Reveree, a diagnostic framework that scores an LLM RE agent's trajectory at three tiers: solve rate, milestone progress through an eight-stage RE schema, and a behavioral profile of its actions. Comprehension stages are scored by an outcome-blinded LLM judge validated against a human expert; all other stages are verified deterministically. Using Reveree, we evaluate nine frontier models and four prompting strategies on 88 picoCTF and NYU-CTF challenges. We find that the base model dominates performance, whereas prompting strategy is a secondary, model-dependent effect. Surprisingly, larger, newer, or costlier models are not reliably stronger. We also find that failures concentrate at the comprehension stages of the RE process, and that extra budget, persistence, or reasoning effort rescues few of them, pointing to a competence limit rather than a resource limit. Regarding memorization, while models reproduce picoCTF flags from challenge descriptions alone, NYU-CTF shows minimal measurable recall, and most solves survive surface perturbation, indicating that genuine analysis coexists with memorization. We release Reveree to the community.

cs.CR↗
arXiv2026

SlideBank: A Persistent Hierarchical Evidence Bank for Consistent Whole-Slide Reasoning

Whole-slide images (WSIs) are challenging for vision-language reasoning because diagnostically relevant morphology is sparse, heterogeneous, and distributed across gigapixel-scale images and multiple spatial resolutions. Existing WSI models and pathology agents can aggregate slide features or actively acquire evidence, but the information retained after exploration is often difficult to access semantically while preserving its connection to the original visual evidence. We introduce SlideBank, a training-free framework that represents each WSI as a persistent, concept-indexed, and spatially grounded evidence bank. SlideBank performs question-independent coarse-to-fine exploration to identify informative regions and multi-scale views, converts them into explicit morphological observations, and grounds pathology signals to their supporting patches and WSI coordinates. At inference time, questions are routed to relevant signals and evidence scales, and the linked global, regional, and patch evidence is integrated through confidence-based cross-level consensus. Experiments on WSI-VQA and SlideBench-BCNB show that with Patho-R1, SlideBank reaches 52.77% on WSI-VQA and with Quilt-LLaVA, it reaches 50.92% average accuracy on SlideBench-BCNB, while structured signal-guided retrieval consistently outperforms random evidence sampling. Reusing the same bank across repeated queries further achieves over 99% rephrasing consistency and substantially reduces amortized inference cost through persistent evidence reuse.

cs.AI↗
arXiv2026

DPA-I2P: Depth-Guided Projective Alignment for Image-to-Point-Cloud Registration in Autonomous Driving

Image-to-Point Cloud Registration aims to estimate the camera pose of a given image within a 3D scene point cloud, which is a fundamental task in autonomous driving and large-scale outdoor localization. Recent implicit correspondence learning methods have improved registration performance by learning cross-modal alignment in an end-to-end framework, leading to more accurate camera pose estimation. However, due to the inherent modality discrepancy between images and sparse LiDAR point clouds, reliable cross-modal correspondence learning remains challenging. To address this issue, we propose Depth-Guided Projective Alignment for Image-to-Point-Cloud Registration (DPA-I2P). Unlike naive depth or feature concatenation, Ray-Conditioned Metric Depth Encoding (RMDE) and Projection-Consistent Vision Lifting (PVL) exploit depth and visual cues in a structured, geometry-aware manner. In addition, Cross-Modal Query Pruning (CQP) suppresses unreliable queries during early refinement to improve matching stability. Experiments on KITTI and nuScenes demonstrate the effectiveness of the proposed method. On KITTI, DPA-I2P reduces RTE and RRE by 45.0% and 55.6% over the strongest implicit baseline, respectively. On nuScenes, DPA-I2P also improves registration accuracy over the evaluated baselines, suggesting better transferability to different driving scenes.

cs.CV↗
arXiv2026

Optical-NIR Multi-band Photometric Analysis and Characterization of Giant Exoplanets with CPI-C

We present a multi-band photometric approach to characterize giant exoplanets, which represents one of the anticipated core scientific outcomes of Cool Planet Imaging Coronagraph (CPI-C). CPI-C operates with two observational channels covering visible and near-infrared wavelengths, each equipped with four broadband filters. The planet--star flux ratio integrated over each filter bandpass is calculated for photometric analysis. For cool planets observed in the visible bands, the data are primarily used to fit the overall spectral shape and methane-induced modulation, providing sensitivity to metallicity- and cloud-dependent spectral variations while constraining the reflected-light spectral shape and the combined scaling involving planet radius, orbital separation, and orbital phase. In the near-infrared bands, which probe thermal emission, the data help to better constrain fundamental planetary parameters including the effective temperature, radius, surface gravity and mass. For a synthetic giant planet with measurable reflected-light and thermal-emission components, the combined VIS4+NIR4 data provide tighter same-target constraints than either filter set alone, especially for the planet radius and cloud sedimentation parameter. Our simulations incorporate realistic instrument throughput, detector noise, and residual speckle noise. The results demonstrate that the eight-band design spanning visible to near-infrared wavelengths supports reflected-light diagnostics, thermal-emission characterization, and joint optical--NIR analysis of giant exoplanets within CPI-C science observations.

astro-ph.EP↗
arXiv2026

Alternative LISA-TAIJI networks: polarization separation of the stochastic gravitational wave background

Stochastic gravitational-wave backgrounds (SGWBs) provide a unique opportunity to probe both unresolved astrophysical populations and fundamental physics in the early Universe. Future space-based gravitational-wave (GW) detectors, such as LISA and TAIJI, will enable cross-correlation observations that are particularly sensitive to SGWBs and their polarization content. In this work, we investigate the capabilities of two proposed LISA-TAIJI networks, LISA-TAIJIp and LISA-TAIJIm, for detecting and discriminating tensor, vector, and scalar polarization components of isotropic SGWBs. Using the cross-correlation between the two networks, we evaluate their sensitivities and signal-to-noise ratios for SGWBs containing different combinations of polarization sectors. We find that the LISA-TAIJIm configuration, which has a larger relative inclination between the two constellations, exhibits substantially weaker polarization degeneracies and significantly improved performance in separating polarization components, particularly for tensor-vector-scalar backgrounds. Fisher-matrix forecasts for power-law SGWBs show that the two configurations yield broadly comparable constraints on spectral parameters, reflecting the different roles of component separation and parameter estimation. Our results demonstrate that large-angle LISA-TAIJI networks provide a more favorable geometry for model-independent SGWB polarization measurements.

gr-qc↗
arXiv2026

Alloy engineering of excitonic properties in TMD monolayers

We investigate monolayer MoS$_{2x}$Se$_{2(1-x)}$ alloys across the full composition range using optical spectroscopy. We demonstrate continuous tuning of the optical gap over $\sim$0.35 eV, accompanied by a systematic reduction of the B--A exciton splitting, in agreement with density functional theory calculations. Temperature-dependent measurements reveal a progressive increase of the average phonon energy from Se-rich to S-rich alloys that follows a simple reduced-mass scaling model. Polarization-resolved spectroscopy further shows a monotonic increase of the circular polarization from nearly zero in MoSe$_2$ to $\sim$15\% in MoS$_2$ at 78 K. The observed evolution of the polarization is attributed to alloy-induced modifications of the electronic structure that modify bright--dark exciton mixing and the associated valley depolarization. These findings establish alloy engineering as an effective strategy for controlling excitonic properties in TMD monolayers.

cond-mat.mtrl-sci↗
arXiv2026

ComFuse: Fusing Complex Memory-Intensive Subgraphs with Compute-Intensive Kernels For Modern GPU Architectures

Modern deep learning workloads increasingly comprise heterogeneous computation graphs that combine compute-intensive operators with memory-intensive subgraphs. Existing deep learning compilers typically optimize these operator classes separately, creating rigid fusion boundaries that limit cross-operator optimization and on-chip data reuse. We observe that downstream memory-intensive operations can execute concurrently with compute-intensive operators, allowing their execution to be hidden behind computation; however, automatically exploiting this opportunity poses new compilation challenges. In this paper, we present ComFuse, an automated GPU compilation system that employs a novel operator fusion strategy to generate high-performance kernels for complex graph structures comprising compute-intensive operators and dependency-rich, memory-intensive elementwise-reduction subgraphs. ComFuse further supports the fusion of back-to-back GEMM (B2BGEMM) patterns, extending its applicability to more complex compute-memory interaction patterns. Additionally, it automatically lowers high-level tensor subprograms into optimized fused kernels, reducing the need for manual kernel engineering. Experimental results show that the fused kernels generated by ComFuse outperform those produced by TorchInductor across post-norm workloads and various complex computation scenarios, while supporting more flexible fusion patterns.

cs.AR↗
arXiv2026

FlowPilot: Real-Time World-Action Modeling for Agile UAV Navigation

We present FlowPilot, a compact world-action model for real-time onboard UAV navigation from depth. Unlike map-then-optimize pipelines that require local reconstruction or end-to-end policies that lack explicit scene prediction, FlowPilot jointly denoises future depth observations and executable trajectories with flow matching. A dual-stream mixture-of-transformers couples video and action experts through shared attention, allowing future-scene prediction and trajectory generation to inform each other. At deployment, the model runs action-centrically and outputs only a trajectory. To ensure trackability, actions are parameterized as degree-7 Bernstein polynomials: the current state constrains the initial control points, and the network predicts five free control points, yielding C^2-continuous references with closed-form velocity, acceleration and jerk. FlowPilot is trained on a three-level depth pyramid spanning high-throughput simulation, photorealistic simulation, and real onboard data. In closed-loop simulation, it outperforms learning- and optimization-based baselines under increasing clutter and commanded speeds up to 8m/s. On a physical quadrotor, the full perception-to-action pipeline runs in under 18ms on a Jetson Orin NX and reaches 5.5m/s in cluttered indoor and forest environments using only onboard sensing and computation.

cs.RO↗
arXiv2026

AeroAct: Action-Centered World-Action Models for Language-Conditioned Quadrotor Flight

Language-conditioned quadrotor flight requires a policy to ground semantic goals, anticipate the visual consequences of ego-motion, and output control references that remain smooth and dynamically executable under rapidly changing first-person views. Existing aerial vision-language navigation and vision-language-action methods commonly use discrete actions, high-level waypoints, or instantaneous velocity commands, which provide limited supervision about how flight actions change future observations. We present AeroAct, an action-centered world-action model (WAM) for quadrotor navigation. To the best of our knowledge, AeroAct is the first WAM instantiated and demonstrated for real-world aerial flight. The model adapts a pretrained video diffusion Transformer to predict local trajectory-action chunks from egocentric visual history, proprioception, and language. Future first-person frames are used during training as dense consequence supervision, while deployment directly decodes actions without generating future video. To obtain aligned visual, state, language, and dynamically feasible action data, we build a DiffAero-based pipeline with complementary Isaac Lab and 3D Gaussian splatting renderers. We further introduce a low-cost handheld collection device that couples camera observations with motion estimates to recreate flight-like egocentric trajectories, and a self-guidance procedure that improves temporal consistency across overlapping trajectory chunks. Closed-loop simulation and real-world experiments show that temporal visual context improves target tracking and object-search performance, and that WAM-based policies can be executed on a physical quadrotor.

cs.RO↗
arXiv2026

Full-Pipeline Inference Optimization for MiMo-V2.5 Series: Pushing Hybrid SWA Efficiency to the Limit

We present a full-pipeline inference optimization for the MiMo-V2.5 model family, which combines Hybrid Sliding Window Attention (Hybrid SWA), sparse Mixture-of-Experts (MoE), and multimodal encoders. While Hybrid SWA can ideally reduce both attention compute and KVCache storage significantly compared to Full Attention, realizing these gains in production requires substantial engineering effort. We systematically optimize the KVCache system with layerwise prefetch, SWA-aware prefix cache trees, and specialized placement strategies, achieving strict $O(W)$ SWA storage and high cache hit rates. We further build GCache, a high-performance distributed cache infrastructure with RDMA-optimized networking, and develop a KVCache-affinity router to reduce computation while preserving load balancing. We also optimize for multimodal inputs, including GPU image preprocessing, parallel video decoding, and multimodal cache sharing. Together, these optimizations constitute the first large-scale LLM serving system in production that efficiently covers the Hybrid SWA + MoE + multimodal composite architecture.

cs.AR↗
arXiv2026

FSDC-DETR: A Frequency-Spatial Domain Collaborative DETR for Small Object Detection

Small object detection (SOD) remains a challenging task in real-world applications. Despite recent advances, existing detectors remain limited by rigid processing that entangle spatial aggregation with implicit frequency aliasing and truncation, leading to inadequate preservation of high-frequency components for SOD. To tackle these limitations, we propose a Frequency-Spatial Domain Collaborative Detection Transformer (FSDC-DETR), a novel collaborative framework that explicitly models complementary spatial and frequency representations. Specifically, we first introduce Dual-Branch Frequency-Spatial Adaptive Fusion (DBFSAF) to enhance frequency diversity and adaptively capture frequency-spatial domain discriminative representations. Building on these representations, a frequency-spatial interaction scheme is further explored within the hybrid encoder to enable progressive feature propagation to the decoder. In particular, structure-aware frequency-spatial aggregation is achieved through Shunt Frequency-Spatial Feature Fusion (SFS-FF), establishing bidirectional interaction and progressive cross-scale propagation between frequency and spatial representations for coherent discriminative modeling. Meanwhile, informative high-frequency responses are preserved during scale transitions through Frequency-Spatial Dynamic Downsampling (FSD-Down), thereby minimizing frequency degradation throughout multi-scale fusion for the precise SOD. Experimental results demonstrate that FSDC-DETR achieves state-of-the-art performance, improving AP by 6.4 on VisDrone-DET2019 and 6.6 on AITODv2, with gains of 6.8 and 6.9 AP for small objects. The code is available at github.com/nevereverinsomnia/FSDC-DETR.

cs.CV↗
arXiv2026

Data-driven Kernel-based Predictive Control with Stability and Robustness Guarantees

In this paper, we provide a theoretical analysis of the closed-loop properties of a data-driven kernel-based predictive control (DDKPC) scheme developed solely from input-output data. The proposed formulation integrates a robust data-driven predictive control framework with a multi-step predictor for nonlinear systems constructed via kernel-based methods. This predictor implicitly captures the system's nonlinear behavior using the representer theorem. For the nominal case with noise-free data, we prove that the DDKPC scheme guarantees recursive feasibility and closed-loop stability, provided that the prediction horizon is sufficiently long and the kernel representation error is sufficiently small. To facilitate real-time implementation, we introduce a penalty relaxation formulation to alleviate the computational burden inherently caused by nonconvex implicit constraints. Furthermore, the framework is robustified against measurement noise by aggregating the representation mismatch and the bounded noise into a unified uncertainty bound. Finally, we extend the DDKPC framework to slowly time-varying nonlinear systems by periodically reconstructing the kernel predictor from a fixed-budget online dictionary managed by the approximate linear dependency (ALD) criterion. Under suitable conditions on the rate of variation of the input-output evolution and the online prediction error, recursive feasibility and practical closed-loop stability are preserved. The effectiveness of the proposed approach is illustrated through numerical examples.

eess.SY↗
arXiv2026

Words Speak Louder Than Code: Investigating Cognitive Heuristics in LLM-Based Code Vulnerability Detection

Researchers and practitioners increasingly apply Large Language Models (LLMs) for automated vulnerability detection. Recent work has shown that LLMs are susceptible to the same cognitive heuristics that bias human judgment. Yet, no work has investigated whether these heuristics affect a model's assessment of code vulnerabilities. In this paper, we present the first systematic exploration of cognitive heuristics in LLM-driven code vulnerability detection. We introduce a controlled framework that holds the code fixed and only varies the surrounding context to trigger three cognitive heuristics: the halo effect through author attribution, the framing effect through task objectives and consequences, and the anchoring effect through prior analysis results. Within this framework, we evaluate eight LLMs across three programming languages and perform both quantitative and code-level analyses. Our findings demonstrate that all evaluated models are susceptible to these heuristics. Cross-model average susceptibility is highest for framing at 33.2%, followed by anchoring at 23.5% and halo at 18.4%. Code-level analysis reveals that vulnerabilities that require semantic reasoning for detection are more susceptible to cognitive heuristics than those identifiable through pattern matching. Furthermore, models often change their verdict from safe to vulnerable based on the cognitive condition, without accurately identifying the actual vulnerability. To highlight the practical impact, we demonstrate a proof-of-concept black-box cognitive attack that can suppress up to 97% of previously detected vulnerabilities. These findings indicate that cognitive susceptibility is a consistent and exploitable property of LLM-based vulnerability detection.

cs.CR↗
arXiv2026

AgentX: Towards Agent-Driven Self-Iteration of Industrial Recommender Systems

Recommendation algorithm iteration is moving from an artisanal, engineer-bound process toward an industrialized research loop, but this transition remains blocked by a structural execution bottleneck: the idea-to-launch cycle still depends on human engineers to generate hypotheses, modify production code, launch A/B experiments, and attribute online results. Innovation therefore scales linearly with headcount rather than compounding with evidence, compute, and accumulated experimental knowledge. We present AgentX, a production-deployed multi-agent system that fundamentally restructures this production function. AgentX operates as a self-evolving development engine: it autonomously generates, implements, evaluates, and learns from recommendation experiments at a scale and pace that no manual workflow can sustain. The system orchestrates four tightly coupled stages in a closed loop. A Brainstorm Agent synthesizes evidence from historical experiments, system architecture, data analysis, and external research into ranked, executable proposals. A Developing Agent translates each proposal into production-ready code through repository-grounded generation and multi-dimensional reliability verification. An Evaluation Agent conducts safe online rollout with guardrail-vetoed A/B judgment, converting both successes and failures into structured knowledge assets. A Harness Evolution layer (SGPO) then distills execution trajectories into semantic-gradient updates that continuously sharpen the agents themselves -- making the system not merely automated, but self-improving.

cs.AI↗
arXiv2026

Data-Driven Robust MPC for Unknown Nonlinear Systems via Set-Membership Learning

Data-driven model predictive control (MPC) has become an attractive approach for controlling unknown systems, especially when data are corrupted by noise. However, most existing data-driven MPC methods focus on linear systems, and little attention has been given to nonlinear dynamics under disturbances. To fill this gap, we propose a robust data-driven min-max MPC scheme for unknown nonlinear systems with process disturbances. We represent the unknown nonlinear dynamics using vector fields built from a dictionary of basis functions, yielding an equivalent linear form with unknown matrices. These unknown matrices are characterized by a set-membership representation derived from noisy input-state data. Using this uncertainty description, we formulate a min-max MPC problem. Two online scenarios are studied: i) when state measurements are noise-free, and, ii) when they are corrupted by process disturbance. For each case, we derive a Lyapunov-based semidefinite program (SDP) to compute a stabilizing state-feedback controller. The resulting schemes are shown to guarantee recursive feasibility and either exponential or robust stability of the closed-loop system depending on whether there is process disturbance. Simulation studies on benchmark examples illustrate the effectiveness and competitive performance of the proposed approach compared to existing data-driven and model-based controllers.

eess.SY↗
arXiv2026

Robust Data-Driven Nash Equilibrium Seeking under Partial-Decision Information

This paper presents a data-driven framework for decentralized Nash equilibrium (NE) seeking in multi-agent systems with unknown linear dynamics subject to exogenous disturbances, operating under partial-decision information (where agents lack direct access to the decisions of all others) and equality constraints. The proposed framework integrates an NE model, a distributed communication protocol, an internal model for disturbance rejection, and a data-driven stabilization strategy. By reformulating the problem as a cooperative output regulation problem, we synthesize controllers directly from noisy input-state data via semi-definite programs (SDPs), providing formal guarantees for closed-loop stability and asymptotic convergence to the NE. The approach is further extended to a class of nonlinear systems with constant disturbances by leveraging integral control and describing nonlinearities via quadratic constraints. Numerical simulations involving unmanned aerial vehicle networks and a rotary-wing aerial vehicle formation validate the efficacy and robustness of the proposed method.

eess.SY↗
arXiv2026

Empowering Polymeric Materials Discovery by Artificial Intelligence

Polymeric materials underpin modern technologies spanning energy storage, microelectronics, healthcare and sustainable manufacturing. Yet their rational design remains exceptionally challenging because material performance emerges from complex interactions among molecular composition, chain architecture, processing history and hierarchical structural evolution across multiple length and time scales. Consequently, polymer research has long relied on labor-intensive experimentation and fragmented modeling approaches, limiting both mechanistic understanding and innovation efficiency. Recent advances in data infrastructure, machine learning, large artificial intelligence (AI) models and laboratory automation are beginning to reshape this landscape. Rather than functioning as isolated tools, polymer databases, predictive models, AI agents and automated laboratories are increasingly converging into interconnected discovery ecosystems. As a result, the central challenge is shifting from improving predictive accuracy alone to enabling reliable decision-making, adaptive learning and seamless integration across computation, experimentation and scientific reasoning. We argue that polymer science is entering an era of autonomous discovery, in which data, simulation, reasoning and experimentation operate within self-improving feedback loops that continuously generate hypotheses, design materials, execute experiments and refine predictive models. By unifying molecular design, process optimization, experimental validation and industrial translation, such autonomous ecosystems establish a more predictive, reproducible and scalable paradigm for polymer innovation, fundamentally transforming how polymer research is conducted.

physics.chem-ph↗
arXiv2026

Data-Driven Frequency-Selective Output Regulation of Nonlinear Systems under Almost Periodic Exosignals

This paper studies output regulation for a class of unknown continuous-time nonlinear systems driven by almost periodic exosignals. The plant dynamics are assumed to be linearly parameterized over a prescribed nonlinear dictionary, while all coefficient matrices in the plant, input channel, output map, and exosignal channel are unknown. Since the plant model is unavailable, exact nonlinear output regulation would generally require model identification followed by the solution of nonlinear regulator equations. To avoid these steps, we pursue a frequency-selective regulation objective: the steady-state regulation error is allowed to be almost periodic, but its Fourier-Bohr coefficients at prescribed exosystem frequencies are guaranteed to vanish, and the residual error energy is explicitly bounded. To this end, a p-copy internal model is embedded in a dynamic controller, yielding an augmented nonlinear system whose unknown constant matrices are represented directly by measured data. A noise-robust semidefinite program is derived to synthesize the controller gain without model identification and without measuring the exosignal amplitudes or phases. The resulting closed-loop vector field is made exponentially contractive on a prescribed operating set, which implies the existence and uniqueness of a bounded and attracting trajectory. By combining contraction theory with Fourier-Bohr analysis, we prove that this steady-state trajectory is almost periodic, that the embedded-frequency components of the regulation error are eliminated, and that the unmodeled spectral components satisfy a Parseval-type time-averaged energy bound. Numerical and physics-based simulations on a quadrotor with a cable-suspended payload illustrate the effectiveness of the proposed data-driven internal-model design.

eess.SY↗