arXiv subjects

Zhenhao Li

Publications and source records attributed to Zhenhao Li.

At least 19 recordsLinked to original sources

arXiv2026

Continuity equation on metric spaces via measure-valued derivations and BV-Wasserstein curves

We introduce a notion of continuity equation on metric spaces that is capable of describing curves of probability measures which are absolutely continuous, and more generally of bounded variation (BV), with respect to the 1-Wasserstein distance. This continuity equation is based on a notion of measure-valued derivations, whose basic theory is also developed in this paper. On $\mathbb{R}^n$, our formulation is consistent with the continuity equation with singular flux introduced by Almi--Rossi--Savar\'e (arXiv:2506.15333), including the corresponding notion of minimal solutions. In this work, we characterize BV-curves in the space of probability measures equipped with the (extended) 1-Wasserstein distance as those curves satisfying the continuity equation with a measure-valued derivation of finite mass. To this aim, we extend our previous work (Calc.Var.(2024)63:16) on probabilistic representations on BV-curves and construct from them measure-valued derivations (resp. flux measures) on geodesic metric spaces (resp. on $\mathbb{R}^n$).

math.AP↗

arXiv2026

AgentExecutor: Partial Code Execution via Agentic Context Generation

Executing code snippets is essential for dynamic program analysis, but it remains challenging to execute an arbitrary code snippet due to issues like missing context and incomplete dependencies. Existing approaches to partial code execution, such as LExecutor and Treefix, leverage the power of language models to infer missing information and enable execution. However, they suffer from (i) limited action spaces and feedback, and (ii) rigid optimization strategies, which restrict their effectiveness and efficiency. In this paper, we propose AgentExecutor, a novel multi-agent framework for partial code execution. Our approach introduces a three-phase design: execution environment preparation, dynamic exploration with iterative refinement, and prefix evolution via program synthesis. Supported by the power of LLM agents who can think, act, and get feedback iteratively, AgentExecutor is able to autonomously explore a richer action space, enabling diverse operations such as creating resource files and resolving environment configuration. Furthermore, it adopts adaptive optimization strategies, including coverage-guided context pruning and prefix evolution via program synthesis, to systematically improve the execution quality of partial code. We evaluate AgentExecutor on two widely used datasets comprising Stack Overflow snippets and open-source project code. The results show that AgentExecutor achieves up to 94% and 90% code coverage, outperforming the state-of-the-art approach Treefix by 19.9% and 13.8%, respectively. In addition, AgentExecutor significantly reduces execution time (by up to 80.3%) and cost (by up to 56.6%). These findings demonstrate that AgentExecutor provides an effective and efficient solution for partial code execution.

cs.SE↗

arXiv2026

CARVE: Cross-Slice Anisotropic Reallocation of Visual Evidence for Efficient 3D Medical Volume Understanding

Slice-based MLLMs leverage mature 2D encoders by representing 3D volumes as sequences of 2D slices. However, this slice-wise formulation produces thousands of visual tokens that burden the LLM backbone, many of which capture overlapping visual evidence across adjacent slices. To understand how effectively a growing visual token budget improves performance, we perform scaling analyses on two 3D medical VQA benchmarks and find diminishing returns: cost keeps rising while accuracy saturates, and improving in-plane resolution is more effective than adding slices at comparable budgets. The budget should therefore be allocated more selectively rather than simply enlarged, yet most token compression methods are designed for 2D images or videos, where redundancy arises from spatial layout or temporal motion rather than from near-duplicate content along the depth axis. We present CARVE, a training-free framework that compresses visual tokens prior to LLM inference and casts token reduction as budget-constrained 2.5D allocation. CARVE partitions the depth axis into coherent windows and allocates tokens non-uniformly according to normalized cross-slice evidence. Under a shared budget, CARVE builds spatial anchors on representative slices and retrieves locally varying evidence from the full volume, then merges remaining eligible tokens into nearby anchors within each window. Removing roughly 80% of the visual tokens on Hulu-Med-7B, CARVE leads all compression baselines on every AMOS-MM report-generation metric, with 6.2 points higher retention of full-token quality than the strongest baseline, and preserves 98.1% of full-token performance across three VQA benchmarks.

cs.CV↗

arXiv2026

SWE-NFI: Studying and Benchmarking Coding Agents for Non-Functional Improvements

Although coding agents have achieved impressive performance on correctness-oriented benchmarks, their ability to make behavior-preserving non-functional improvements (NFIs) remains underexplored. In real-world software development, developers continuously improve software quality without changing observable behavior, yet existing benchmarks primarily evaluate functional correctness and provide limited support for assessing these non-functional improvements. In this paper, we present SWE-NFI, a benchmark for evaluating coding agents on NFIs beyond functional correctness. Our benchmark contains 188 tasks constructed from real merged pull requests in open-source Python projects. We operationalize developer-oriented NFIs into 92 executable rules and develop a comprehensive evaluation suite that combines functional correctness testing with rule-based NFI evaluation. We evaluate state-of-the-art commercial and open-source coding agents. Although the best-performing agent achieves a 70.0\% functional correctness rate, all evaluated agents generally fall short of human developers in overall NFI capability. The gap is particularly evident for structural code improvements, where agents' NFI scores range from 0.0 to 1.3, compared with 1.5 for the human reference. Our benchmark and findings provide a reproducible foundation for evaluating and advancing coding agents beyond functional correctness.

cs.SE↗

arXiv2026

Brain-Adapter: A Dual-Stream Vision-Language MIL Framework for Comprehensive 3D CT Diagnosis of Acute Intracranial Pathologies

Automated diagnosis of 3D brain CT scans is essential for critical care, yet it remains challenging due to the heavy reliance on manual annotations and the limited semantic understanding of conventional models. While 2D foundation vision-language models (VLMs) have shown remarkable generalization, effectively transferring their representational power to 3D volumes remains an open problem. In this paper, we propose Brain-Adapter, a novel dual-stream multiple instance learning (MIL) framework that leverages pre-trained 2D biomedical VLMs and raw diagnostic reports for robust scan-level multi-label classification. Specifically, we introduce a Text-Conditioned Attention (TCA) mechanism, utilizing raw diagnostic sentences as semantic queries to dynamically align visual cues with specific disease concepts. Concurrently, a parallel visual MIL stream captures global scan characteristics, supervised by structured labels extracted via a Large Language Model (LLM). To ensure representation coherence, a consistency constraint enforces synergy between the two streams. During inference, an Uncertainty-Aware Refinement (UAR) module dynamically calibrates and fuses these dual-stream predictions to resolve ambiguous cases. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art 3D models and standard MIL approaches. By eliminating the reliance on dense annotations, Brain-Adapter provides a highly scalable and clinically viable solution for 3D acute intracranial pathology analysis.

cs.CV↗

arXiv2026

Harsher on Male? Evaluating LLMs on Gender-Asymmetric Moral Framing Across Diverse Conflict Scenarios

Existing studies on gender bias in LLMs have largely focused on stereotypes, occupational associations, or explicit harmful outputs. In this work, we ask whether LLMs apply consistent response standards to the same negative behavior under matched male-actor and female-actor conditions. We introduce GAMA-Bench, a gender-mirrored benchmark of 1,298 scenarios covering intimate relationship and public social conflicts. It constructs gender-neutral misconduct templates through controlled grids and cross-model review, then compiles them into paired first-person prompts with matched actor-gender and role-reference variations. We further design a structured response-framing protocol to measure how models allocate punishment, empathy, escalation, instruction, and blame. Experiments on 10 representative LLMs reveal a consistent male-disadvantaging asymmetry: male actors receive more punitive, escalatory, and blame-centered framing, whereas female actors receive more therapeutic and empathy-oriented framing for the same misconduct. Further analyses show that this pattern persists across model families, scenario tracks, model scale, and explicit thinking-style reasoning. The official code is available at https://github.com/xufeiqiong/GAMA-Bench.

cs.CL↗

arXiv2026

High-quality Nano-patterning of Oxide Interfaces Using Transferred Gold Mask

Complex oxide interfaces, such as $\mathrm{SrTiO_3}$ and $\mathrm{KTaO_3}$ based heterostructures, host rich correlated phenomena with strong potential for advanced device applications. However, these interfaces are extremely susceptible to contamination and defect formation during nanofabrication, which often compromises device performance. Here, we present a solvent-free method for patterning oxide interfaces by employing high-resolution transferable thin metal masks in conjunction with oxygen-enriched $\mathrm{Ar^+}$ ion milling, which enables a clean and well-controlled nanofabrication process. Transport measurements demonstrate that the fabricated devices preserve their intrinsic properties, including high carrier mobilities, with negligible degradation compared to the pristine interfaces. This technique offers a convenient and robust route for engineering high-performance oxide electronic devices with precisely tailored transport characteristics.

cond-mat.mtrl-sci↗

arXiv2026

Towards Secure Logging: Characterizing and Benchmarking Logging Code Security Issues with LLMs

Logging code plays an important role in software systems by recording key events and behaviors, which are essential for debugging and monitoring. However, insecure logging practices can inadvertently expose sensitive information or enable attacks such as log injection, posing serious threats to system security and privacy. Prior research has examined general defects in logging code, but systematic analysis of logging code security issues remains limited, particularly in leveraging LLMs for detection and repair. In this paper, we derive a comprehensive taxonomy of logging code security issues, encompassing four common issue categories and 10 corresponding patterns. We further construct a benchmark dataset with 101 real-world logging security issue reports that have been manually reviewed and annotated. We then propose an automated framework that incorporates various contextual knowledge to evaluate LLMs' capabilities in detecting and repairing logging security issues. Our experimental results reveal a notable disparity in performance: while LLMs are moderately effective at detecting security issues (e.g., the accuracy ranges from 12.9% to 52.5% on average), they face noticeable challenges in reliably generating correct code repairs. We also find that the issue description alone improves the LLMs' detection accuracy more than the security pattern explanation or a combination of both. Overall, our findings provide actionable insights for practitioners and highlight the potential and limitations of current LLMs for secure logging.

cs.SE↗

arXiv2026

How Adversarial Environments Mislead Agentic AI?

Tool-integrated agents are deployed on the premise that external tools ground their outputs in reality. Yet this very reliance creates a critical attack surface. Current evaluations benchmark capability in benign settings, asking "can the agent use tools correctly" but never "what if the tools lie". We identify this Trust Gap: agents are evaluated for performance, not for skepticism. We formalize this vulnerability as Adversarial Environmental Injection (AEI), a threat model where adversaries compromise tool outputs to deceive agents. AEI constitutes environmental deception: constructing a "fake world" of poisoned search results and fabricated reference networks around unsuspecting agents. We operationalize this via POTEMKIN, a Model Context Protocol (MCP)-compatible harness for plug-and-play robustness testing. We identify two orthogonal attack surfaces: The Illusion (breadth attacks) poison retrieval to induce epistemic drift toward false beliefs, while The Maze (depth attacks) exploit structural traps to cause policy collapse into infinite loops. Across 11,000+ runs on five frontier agents, we find a stark robustness gap: resistance to one attack often increases vulnerability to the other, demonstrating that epistemic and navigational robustness are distinct capabilities.

cs.AI↗

arXiv2026

ReLog: Execution-Aware Logging with Runtime Feedback for LLM-Oriented Debugging

Logging statements are essential for software debugging and maintenance. However, existing approaches to automatic logging generation rely on static analysis and produce statements in a single pass without considering runtime behavior. They are also typically evaluated by similarity to developer-written logs, assuming these logs form an adequate gold standard. This assumption is increasingly limiting in the LLM era, where logs are consumed not only by developers but also by LLMs for downstream tasks. As a result, optimizing logs for human similarity does not necessarily reflect their practical utility. To address these limitations, we introduce ReLog, an iterative logging generation framework guided by runtime feedback. ReLog leverages LLMs to generate, execute, evaluate, and refine logging statements so that runtime logs better support downstream tasks. Instead of comparing against developer-written logs, we evaluate ReLog through downstream debugging tasks, including defect localization and repair. We construct a benchmark based on Defects4J under both direct and indirect debugging settings. Results show that ReLog consistently outperforms all baselines, achieving an F1 score of 0.520 and repairing 97 defects in the direct setting, and the best F1 score of 0.408 in the indirect setting where source code is unavailable. Additional experiments across multiple LLMs demonstrate the generality of the framework, while ablations confirm the importance of iterative refinement and compilation repair. Overall, our work reframes logging as a runtime-guided, task-oriented process and advocates evaluating logs by their downstream utility rather than textual similarity.

cs.SE↗

arXiv2026

Real-time Appearance-based Gaze Estimation for Open Domains

Appearance-based gaze estimation (AGE) has achieved remarkable performance in constrained settings, yet we reveal a significant generalization gap where existing AGE models often fail in practical, unconstrained scenarios, particularly those involving facial wearables and poor lighting conditions. We attribute this failure to two core factors: limited image diversity and inconsistent label fidelity across different datasets, especially along the pitch axis. To address these, we propose a robust AGE framework that enhances generalization without requiring additional human-annotated data. First, we expand the image manifold via an ensemble of augmentation techniques, including synthesis of eyeglasses, masks, and varied lighting. Second, to mitigate the impact of anisotropic inter-dataset label deviation, we reformulate gaze regression as a multi-task learning problem, incorporating multi-view supervised contrastive (SupCon) learning, discretized label classification, and eye-region segmentation as auxiliary objectives. To rigorously validate our approach, we curate new benchmark datasets designed to evaluate gaze robustness under challenging conditions, a dimension largely overlooked by existing evaluation protocols. Our MobileNet-based lightweight model achieves generalization performance competitive with the state-of-the-art (SOTA) UniGaze-H, while utilizing less than 1\% of its parameters, enabling high-fidelity, real-time gaze tracking on mobile devices.

cs.CV↗

arXiv2026

Reconfigurable Oxide Nanoelectronics by Tip-induced Electron Delocalization

Reconfigurable oxide nanoelectronics, enabled by conductive atomic force microscope (cAFM) lithography, have established complex oxide interfaces as a promising platform for quantum engineering that harnesses emergent phenomena for advanced functionalities. However, this cAFM nanofabrication process can only occur in the air, with simultaneous device decay described under the "water-cycle" writing mechanism. These restrictions pose ongoing challenges for device optimization in the quantum regime at mK temperatures. Here, we demonstrate a "waterless" cAFM lithography approach that is compatible with vacuum and cryogenic environments. Through oxygen vacancy engineering at the LaAlO$_3$/SrTiO$_3$ interface, we have achieved nonvolatile and reconfigurable cAFM control of nanoscale interfacial polaron-electron liquid transition at mK temperatures with an ultrafine line resolution of 0.85 nm. Supported by first-principles calculations and drift-diffusion modeling, we show that tip-controlled oxygen vacancy electromigration plays a key role. This advancement bridges reconfigurable device fabrication and concurrent characterization in situ at mK temperatures, and establishes a versatile Hubbard toolbox for engineering programmable quantum phases in correlated oxides.

cond-mat.mes-hall↗

arXiv2025

Few-Shot-Based Modular Image-to-Video Adapter for Diffusion Models

Diffusion models (DMs) have recently achieved impressive photorealism in image and video generation. However, their application to image animation remains limited, even when trained on large-scale datasets. Two primary challenges contribute to this: the high dimensionality of video signals leads to a scarcity of training data, causing DMs to favor memorization over prompt compliance when generating motion; moreover, DMs struggle to generalize to novel motion patterns not present in the training set, and fine-tuning them to learn such patterns, especially using limited training data, is still under-explored. To address these limitations, we propose Modular Image-to-Video Adapter (MIVA), a lightweight sub-network attachable to a pre-trained DM, each designed to capture a single motion pattern and scalable via parallelization. MIVAs can be efficiently trained on approximately ten samples using a single consumer-grade GPU. At inference time, users can specify motion by selecting one or multiple MIVAs, eliminating the need for prompt engineering. Extensive experiments demonstrate that MIVA enables more precise motion control while maintaining, or even surpassing, the generation quality of models trained on significantly larger datasets.

cs.CV↗

arXiv2025

LLM4Perf: Large Language Models Are Effective Samplers for Multi-Objective Performance Modeling

The performance of modern software systems is critically dependent on their complex configuration options. Building accurate performance models to navigate this vast space requires effective sampling strategies, yet existing methods often struggle with multi-objective optimization and cannot leverage semantic information from documentation. The recent success of Large Language Models (LLMs) motivates the central question of this work: Can LLMs serve as effective samplers for multi-objective performance modeling? To explore this, we present a comprehensive empirical study investigating the capabilities and characteristics of LLM-driven sampling. We design and implement LLM4Perf, a feedback-based framework, and use it to systematically evaluate the LLM-guided sampling process across four highly configurable, real-world systems. Our study reveals that the LLM-guided approach outperforms traditional baselines in most cases. Quantitatively, LLM4Perf achieves the best performance in nearly 68.8% (77 out of 112) of all evaluation scenarios, demonstrating its superior effectiveness. We find this effectiveness stems from the LLM's dual capabilities of configuration space pruning and feedback-driven strategy refinement. The effectiveness of this pruning is further validated by the fact that it also improves the performance of the baseline methods in nearly 91.5% (410 out of 448) of cases. Furthermore, we show how the LLM choices for each component and hyperparameters within LLM4Perf affect its effectiveness. Overall, this paper provides strong evidence for the effectiveness of LLMs in performance engineering and offers concrete insights into the mechanisms that drive their success.

cs.SE↗

arXiv2025

Neural Factorization-based Bearing Fault Diagnosis

This paper studies the key problems of bearing fault diagnosis of high-speed train. As the core component of the train operation system, the health of bearings is directly related to the safety of train operation. The traditional diagnostic methods are facing the challenge of insufficient diagnostic accuracy under complex conditions. To solve these problems, we propose a novel Neural Factorization-based Classification (NFC) framework for bearing fault diagnosis. It is built on two core idea: 1) Embedding vibration time series into multiple mode-wise latent feature vectors to capture diverse fault-related patterns; 2) Leveraging neural factorization principles to fuse these vectors into a unified vibration representation. This design enables effective mining of complex latent fault characteristics from raw time-series data. We further instantiate the framework with two models CP-NFC and Tucker-NFC based on CP and Tucker fusion schemes, respectively. Experimental results show that both models achieve superior diagnostic performance compared with traditional machine learning methods. The comparative analysis provides valuable empirical evidence and practical guidance for selecting effective diagnostic strategies in high-speed train bearing monitoring.

cs.LG↗

arXiv2025

Synthetic approaches to Ricci flows

We review different notions of synthetic Ricci flow that apply to time-dependent families of metric measure spaces and which are based on properties of the heat flow, ideas from optimal transport, and the asymptotic behaviour of volumes. Each notion equivalently characterises (weighted) Ricci flow for smooth families of weighted Riemannian manifolds. We discuss the features of the different notions on various examples.

math.DG↗

arXiv2025

Evaluating Software Process Models for Multi-Agent Class-Level Code Generation

Modern software systems require code that is not only functional but also maintainable and well-structured. Although Large Language Models (LLMs) are increasingly used to automate software development, most studies focus on isolated, single-agent function-level generation. This work examines how process structure and role specialization shape multi-agent LLM workflows for class-level code generation. We simulate a Waterfall-style development cycle covering Requirement, Design, Implementation, and Testing using three LLMs (GPT-4o-mini, DeepSeek-Chat, and Claude-3.5-Haiku) on 100 Python tasks from the ClassEval benchmark. Our findings show that multi-agent workflows reorganize, rather than consistently enhance, model performance. Waterfall-style collaboration produces cleaner and more maintainable code but often reduces functional correctness (-37.8\% for GPT-4o-mini and -39.8\% for DeepSeek-Chat), with Claude-3.5-Haiku as a notable exception (+9.5\%). Importantly, process constraints shift failure characteristics: structural issues such as missing code decrease, while semantic and validation errors become more frequent. Among all stages, Testing exerts the strongest influence by improving verification coverage but also introducing new reasoning failures, whereas Requirement and Design have comparatively modest effects. Overall, this study provides empirical evidence that software process structure fundamentally alters how LLMs reason, collaborate, and fail, revealing inherent trade-offs between rigid workflow discipline and flexible problem-solving in multi-agent code generation.

cs.SE↗

arXiv2025

SecureVibeBench: Benchmarking Secure Vibe Coding of AI Agents via Reconstructing Vulnerability-Introducing Scenarios

Large language model-powered code agents are rapidly transforming software engineering, yet the security risks of their generated code have become a critical concern. Existing benchmarks have provided valuable insights, but they fail to capture scenarios in which vulnerabilities are actually introduced by human developers, making fair comparisons between humans and agents infeasible. We therefore introduce SecureVibeBench, a benchmark of 105 C/C++ secure coding tasks sourced from 41 projects in OSS-Fuzz for code agents. SecureVibeBench has the following features: (i) realistic task settings that require multi-file edits in large repositories, (ii)~aligned contexts based on real-world open-source vulnerabilities with precisely identified vulnerability introduction points, and (iii) comprehensive evaluation that combines functionality testing and security checking with both static and dynamic oracles. We evaluate 5 popular code agents like OpenHands, supported by 5 LLMs (e.g., Claude sonnet 4.5) on SecureVibeBench. Results show that current agents struggle to produce both correct and secure code, as even the best-performing one, produces merely 23.8\% correct and secure solutions on SecureVibeBench. Our code and data are on https://github.com/iCSawyer/SecureVibeBench.

cs.SE↗