arXiv subjects

Lei Xue

Publications and source records attributed to Lei Xue.

At least 19 recordsLinked to original sources

arXiv2026

Error-Aware Reverse Auction Mechanism for Large Language Model Routing

Routing each query to a cost-effective large language model (LLM) is critical for balancing quality and cost, yet most routers rely on a centralized task center to predict model performance, creating an information-risk mismatch and a scalability bottleneck as the model pool grows. We propose a market-based routing paradigm that shifts ex-ante prediction to LLM providers via a reverse auction, where providers bid with self-predicted success probabilities and execution costs. To account for inherently noisy provider predictions and center evaluations, we introduce the \textit{\textbf{E}rror-\textbf{A}ware \textbf{R}everse \textbf{A}uction \textbf{M}echanism} (EA-RAM), which explicitly models this inherent Dual Error. We prove that EA-RAM is Bayesian incentive compatible and individually rational under the Dual Error, establish sufficient conditions for center rationality, and derive an explicit welfare-loss bound. We further identify robustness effects: opposite-signed errors can cancel, vanishing-tail link functions (e.g., logistic) stabilize clear-cut cases via saturation, and extra noise smooths belief maps, reducing the gains from marginal manipulation. Experiments on simulations and real-world benchmarks show that EA-RAM is robust to the Dual Error and achieves a better cost--performance Pareto frontier than centralized baselines, with additional gains when providers contribute local information, validating its practical effectiveness.

cs.GT↗

arXiv2026

ERSkill: Evolving for Skill-Guided Adaptive Memory Retrieval

While Large Language Model (LLM) agents increasingly rely on long-term memory for persistent interactions, the retrieval mechanisms governing this memory are rarely treated as evolvable components. This static approach limits performance on heterogeneous memory queries, which often demand diverse evidence construction strategies. To address this, we introduce \textbf{ERSkill}, a retrieval-centric framework for self-evolving, skill-guided memory access. ERSkill compiles interaction histories into a structured memory store and represents retrieval behaviors as executable skills composed of fundamental primitives. At inference time, a trained router dynamically matches each query to the optimal skill to construct tailored evidence for answer generation. To enable continuous improvement, ERSkill co-evolves the skill set and the router during training. It employs an experience trie to efficiently record explored retrieval paths, alongside a double-frontier mechanism that safely decouples the expansion of new skill capabilities from stable, router-facing deployment. Experiments across multiple agent memory benchmarks demonstrate that ERSkill substantially outperforms strong non-evolving and self-evolving baselines. Notably, it improves the overall average across F1, BLEU-1, and LLM-judge scores by 31.3\% with Qwen3-Next-80B-A3B-Instruct and by 28.1\% with GPT-5.4-nano.

cs.CL↗

arXiv2026

DreamGuard: Efficient Runtime Guardrail for LLM Agents via Risk-Aware World Model

As large language model (LLM) agents increasingly invoke external tools and interact with real-world systems, unsafe actions may cause irreversible consequences on external states, user data, and downstream services. Recent runtime guardrails mitigate such risks by checking proposed actions before execution, but many remain reactive: they primarily assess the apparent safety of the current action, lacking an explicit model of how risk evolves across the trajectory. This limitation creates a critical blind spot for long-horizon risks, where individually benign-looking actions can gradually drift the agent toward hazardous states. In response, we propose DreamGuard, a proactive guardrail for LLM agents built around a risk-aware world model. The world model maintains a compact recurrent latent state over the trajectory and predicts future latent states from which DreamGuard derives immediate-hazard and prefix-risk evidence. It then fuses these multi-horizon signals into intervention decisions before execution. Experiments across four benchmarks and an online guardrail evaluation show that DreamGuard outperforms generic, reactive, and proactive guardrail baselines, achieves the best safety-utility trade-off among evaluated guardrails, and maintains an average end-to-end latency of 25 ms per call.

cs.AI↗

arXiv2026

Any Proof of Polynomial Hirsch Must be Completely Incoherent

In 1992, Billera and Sturmfels introduced coherent monotone paths on polytopes as part of their description of the fiber polytope construction, and later in 1994 showed with Kapranov that these coherent monotone paths capture the topology of the space of all monotone paths, paths from a minimum to a maximum, in the directed graph of a polytope with orientation induced by a linear function. Those results motivate the following analog of the polynomial Hirsch conjecture: Does there always exist a coherent monotone path of polynomial length on a polytope for any choice of orientation induced by a linear function? We show this is not the case by exhibiting a family of polytopes and corresponding linear functions for which every coherent monotone path is exponentially long. As applications, we strengthen longstanding results pertaining to lower bounds for the shadow simplex method, geometric transversals in discrete geometry, and parametric linear optimization.

math.CO↗

arXiv2026

Dual Lattice Functions of Polytopes

We define the dual lattice function of a rational polytope $P$ via the discrete Laplace transform of the exponential of its support function. This definition is a discrete analogue of the dual volume function of a polytope that the authors studied in previous work. We show that the dual lattice function is valuative, and by multiplying with the torus form, it becomes the canonical form of the exponential polytope $\mathrm{exp}(P)$ as a positive geometry. This result suggests the study of the class of toric polytopes, which are certain semialgebraic subsets of projective toric varieties. Our work is a first step towards discretization of positive geometries in the simplest case of polytopes.

math.CO↗

arXiv2026

SMH-Bench: Benchmarking LLM Agents for Environment-Grounded Reasoning and Action in Smart Homes

Smart homes are evolving toward complex state-dependent living environments, requiring Large Language Models (LLMs) to reason over user intent, preferences, and multi-device interactions. However, existing smart-home benchmarks often focus on static instruction-to-API mapping or limited simulations, failing to evaluate whether LLMs can reason, interact, and act reliably in realistic household scenarios. To address these limitations, we introduce SMH-Bench, a comprehensive benchmark for evaluating LLMs in smart-home environments. Built upon HomeEnv, an executable and verifiable smart-home simulator, SMH-Bench contains 1,100 high-quality tasks spanning 7 categories and 22 fine-grained subcategories. It further stratifies tasks across simple, medium and complex homes, ranging from small apartments to dense multi-room environments with 135 devices. Experiments show that although frontier LLMs achieve strong performance on explicit control and query tasks, they still exhibit significant weaknesses in automation task scheduling, ambiguity handling and personalized reasoning, especially as home complexity increases. We hope SMH-Bench will facilitate the development of more reliable, context-aware, and practically deployable smart-home agents.

cs.AI↗

arXiv2026

HomeFlow: A Data Flywheel for Smart Home Agent Training with Verifiable Simulation

Large language model agents are moving beyond text-only interaction toward physical-world control, with smart homes as a representative domain. Real domestic interaction requires understanding ambiguous intents, operating in dynamic environments, and performing multi-turn reasoning. However, existing methods struggle to generate high-quality training data for smart home agents. We propose HomeFlow, a verifiable data flywheel for this domain. HomeFlow uses HomeEnv as a unified simulation environment and HomeMaker to procedurally generate diverse home settings. Subsequently, Blueprint compiles open-ended user intents into executable state-based success conditions, while MCTS-Flow synthesizes diverse, verifiable multi-turn trajectories through environment-guided tree search. We then optimize the agents via supervised fine-tuning and step-wise RLVE, which facilitates iterative improvement through authentic physical feedback. We further construct SmartHome-Bench to evaluate the agent across various smart home tasks. On this benchmark, HomeFlow-RL-4B and HomeFlow-RL-8B achieve task success rates of 84.60% and 87.03%. It is worth noting that HomeFlow-RL-8B even surpasses the leading GPT-5.5 by 1.23 percentage points.

cs.AI↗

arXiv2026

ICON: Intent-Context Coupling for Efficient Multi-Turn Jailbreak Attack

Multi-turn jailbreak attacks have emerged as a critical threat to Large Language Models (LLMs), bypassing safety mechanisms by progressively constructing adversarial contexts from scratch and incrementally refining prompts. However, existing methods suffer from the inefficiency of incremental context construction that requires step-by-step LLM interaction, and often stagnate in suboptimal regions due to surface-level optimization. In this paper, we characterize the Intent-Context Coupling phenomenon, revealing that LLM safety constraints are significantly relaxed when a malicious intent is coupled with a semantically congruent context pattern. Driven by this insight, we propose ICON, an automated multi-turn jailbreak framework that efficiently constructs an authoritative-style context via prior-guided semantic routing. Specifically, ICON first routes the malicious intent to a congruent context pattern (e.g., Scientific Research) and instantiates it into an attack prompt sequence. This sequence progressively builds the authoritative-style context and ultimately elicits prohibited content. In addition, ICON incorporates a Hierarchical Optimization Strategy that combines local prompt refinement with global context switching, preventing the attack from stagnating in ineffective contexts. Experimental results across eight SOTA LLMs demonstrate the effectiveness of ICON, achieving a state-of-the-art average Attack Success Rate (ASR) of 97.1\%. Code is available at https://github.com/xwlin-roy/ICON.

cs.CR↗

arXiv2025

Assessing the Capability of Android Dynamic Analysis Tools to Combat Anti-Runtime Analysis Techniques

As the dominant mobile operating system, Android continues to attract a substantial influx of new applications each year. However, this growth is accompanied by increased attention from malicious actors, resulting in a significant rise in security threats to the Android ecosystem. Among these threats, the adoption of Anti-Runtime Analysis (ARA) techniques by malicious applications poses a serious challenge, as it hinders security professionals from effectively analyzing malicious behaviors using dynamic analysis tools. ARA technologies are designed to prevent the dynamic examination of applications, thus complicating efforts to ensure platform security. This paper presents a comprehensive empirical study that assesses the ability of widely-used Android dynamic analysis tools to bypass various ARA techniques. Our findings reveal a critical gap in the effectiveness of existing dynamic analysis tools to counter ARA mechanisms, highlighting an urgent need for more robust solutions. This work provides valuable insights into the limitations of existing tools and highlights the need for improved methods to counteract ARA technologies, thus advancing the field of software security and dynamic analysis.

cs.SE↗

arXiv2025

Decentralized Rank Scheduling for Energy-Constrained Multi-Task Federated Fine-Tuning in Edge-Assisted IoV Networks

Large-scale Internet of Vehicles (IoV) deployments increasingly demand the on-device adaptation of foundation models to support diverse, mission-critical perception tasks. While federated fine-tuning offers a promising solution for efficient model specialization, existing approaches often struggle to reconcile the inherent conflict between stringent global energy budgets, heterogeneous task demands, and the high volatility of vehicular network connectivity. In this work, we introduce a hierarchical, adaptive framework that decouples multi-task fine-tuning into two interdependent optimization phases. First, we implement a feedback-loop mechanism at the infrastructure level that dynamically redistributes global energy budgets across concurrent tasks based on real-time convergence dynamics and resource utilization. Second, at the vehicle level, we formulate intra-task rank selection as an energy-constrained online learning problem, solved via a novel primal-dual bandit algorithm, UCB-DUAL, which provides theoretical guarantees on sublinear regret. Our approach effectively internalizes global energy constraints into local decision-making, allowing vehicles to autonomously navigate the complex trade-off between model accuracy, latency, and power consumption. Empirical evaluations using a large-scale IoV simulator, driven by real-world trajectory data, confirm that our proposed method significantly outperforms current federated fine-tuning baselines, offering a robust and scalable solution for resource-constrained vehicular intelligence.

cs.LG↗

arXiv2025

Lorica: A Synergistic Fine-Tuning Framework for Advancing Personalized Adversarial Robustness

The growing use of large pre-trained models in edge computing has made model inference on mobile clients both feasible and popular. Yet these devices remain vulnerable to adversarial attacks, threatening model robustness and security. Federated adversarial training (FAT) offers a promising solution by enhancing robustness while preserving client privacy. However, FAT often yields a generalized global model that struggles with heterogeneous client data, leading to limited personalization and significant communication overhead. In this paper, we propose \textit{Lorica}, a personalized synergistic adversarial training framework that delivers customized defense models through a two-phase process. In Phase 1, \textit{Lorica} applies LoRA-FA for local adversarial fine-tuning, enabling personalized robustness while reducing communication by uploading only LoRA-FA parameters. In Phase 2, a forward-gating selection strategy improves benign accuracy, further refining the personalized model. This yields tailored defense models that effectively balance robustness and accuracy. Extensive experiments on benchmark datasets demonstrate that \textit{Lorica} can achieve up to 68$\times$ improvements in communication efficiency compared to state-of-the-art algorithms, while achieving up to 29.9\% and 52.2\% enhancements in adversarial robustness and benign accuracy, respectively.

cs.CR↗

arXiv2025

Vehicular Intrusion Detection System for Controller Area Network: A Comprehensive Survey and Evaluation

The progress of automotive technologies has made cybersecurity a crucial focus, leading to various cyber attacks. These attacks primarily target the Controller Area Network (CAN) and specialized Electronic Control Units (ECUs). In order to mitigate these attacks and bolster the security of vehicular systems, numerous defense solutions have been proposed.These solutions aim to detect diverse forms of vehicular attacks. However, the practical implementation of these solutions still presents certain limitations and challenges. In light of these circumstances, this paper undertakes a thorough examination of existing vehicular attacks and defense strategies employed against the CAN and ECUs. The objective is to provide valuable insights and inform the future design of Vehicular Intrusion Detection Systems (VIDS). The findings of our investigation reveal that the examined VIDS primarily concentrate on particular categories of attacks, neglecting the broader spectrum of potential threats. Moreover, we provide a comprehensive overview of the significant challenges encountered in implementing a robust and feasible VIDS. Additionally, we put forth several defense recommendations based on our study findings, aiming to inform and guide the future design of VIDS in the context of vehicular security.

cs.CR↗

arXiv2025

Towards Automated Smart Contract Generation: Evaluation, Benchmarking, and Retrieval-Augmented Repair

Smart contracts, predominantly written in Solidity and deployed on blockchains such as Ethereum, are immutable after deployment, making functional correctness critical. However, existing evaluations of Solidity code generation rely largely on surface-level metrics (e.g., BLEU, CrystalBLEU) or manual inspection, which correlate poorly with functional correctness. In contrast to Python, Solidity lacks large-scale, execution-based benchmarks, limiting systematic evaluation of large language models for smart contract development. We introduce SolBench, a comprehensive benchmark and automated testing pipeline for Solidity that emphasizes functional correctness via differential fuzzing. SolBench consists of 28825 functions extracted from 7604 real-world smart contracts collected from Etherscan (genesis-2024), spanning ten application domains. We benchmark 14 diverse LLMs, covering open and closed models, 1.3B-671B parameters, and both general-purpose and code-specialized architectures. The dominant failure mode is missing critical intra-contract information, such as state variables and type definitions. Providing full-contract context improves accuracy but incurs prohibitive inference costs. To address this, we propose Retrieval-Augmented Repair (RAR), a cost-effective framework that integrates execution feedback into code repair. RAR uses compiler and runtime error messages to retrieve only the minimal contract snippets needed to correct a target function, avoiding full-context inference. This significantly reduces input length while improving functional correctness. We further analyze retrieval and repair strategies within RAR, demonstrating consistent gains in accuracy and efficiency. SolBench and RAR enable principled, execution-based evaluation and economical improvement of Solidity code generation. Dataset and code are publicly available at https://github.com/ZaoyuChen/SolBench.

cs.SE↗

arXiv2025

ABFS: Natural Robustness Testing for LLM-based NLP Software

Owing to the exceptional performance of Large Language Models (LLMs) in Natural Language Processing (NLP) tasks, LLM-based NLP software has rapidly gained traction across various domains, such as financial analysis and content moderation. However, these applications frequently exhibit robustness deficiencies, where slight perturbations in input (prompt+example) may lead to erroneous outputs. Current robustness testing methods face two main limitations: (1) low testing effectiveness, limiting the applicability of LLM-based software in safety-critical scenarios, and (2) insufficient naturalness of test cases, reducing the practical value of testing outcomes. To address these issues, this paper proposes ABFS, a straightforward yet effective automated testing method that, for the first time, treats the input prompts and examples as a unified whole for robustness testing. Specifically, ABFS formulates the testing process as a combinatorial optimization problem, employing Best-First Search to identify successful test cases within the perturbation space and designing a novel Adaptive control strategy to enhance test case naturalness. We evaluate the robustness testing performance of ABFS on three datasets across five threat models. On Llama2-13b, the traditional StressTest achieves only a 13.273% success rate, while ABFS attains a success rate of 98.064%, supporting a more comprehensive robustness assessment before software deployment. Compared to baseline methods, ABFS introduces fewer modifications to the original input and consistently generates test cases with superior naturalness. Furthermore, test cases generated by ABFS exhibit stronger transferability and higher testing efficiency, significantly reducing testing costs.

cs.SE↗

arXiv2024

Assessing the Robustness of LLM-based NLP Software via Automated Testing

Benefiting from the advancements in LLMs, NLP software has undergone rapid development. Such software is widely employed in various safety-critical tasks, such as financial sentiment analysis, toxic content moderation, and log generation. Unlike traditional software, LLM-based NLP software relies on prompts and examples as inputs. Given the complexity of LLMs and the unpredictability of real-world inputs, quantitatively assessing the robustness of such software is crucial. However, to the best of our knowledge, no automated robustness testing methods have been specifically designed to evaluate the overall inputs of LLM-based NLP software. To this end, this paper introduces the first AutOmated Robustness Testing frAmework, AORTA, which reconceptualizes the testing process into a combinatorial optimization problem. Existing testing methods designed for DNN-based software can be applied to LLM-based software by AORTA, but their effectiveness is limited. To address this, we propose a novel testing method for LLM-based software within AORTA called Adaptive Beam Search. ABS is tailored for the expansive feature space of LLMs and improves testing effectiveness through an adaptive beam width and the capability for backtracking. We successfully embed 18 test methods in the designed framework AORTA and compared the test validity of ABS with three datasets and five threat models. ABS facilitates a more comprehensive and accurate robustness assessment before software deployment, with an average test success rate of 86.138%. Compared to the currently best-performing baseline PWWS, ABS significantly reduces the computational overhead by up to 3441.895 seconds per successful test case and decreases the number of queries by 218.762 times on average. Furthermore, test cases generated by ABS exhibit greater naturalness and transferability.

cs.SE↗

arXiv2024

A Comprehensive Study on Dark Patterns

As digital interfaces become increasingly prevalent, certain manipulative design elements have emerged that may harm user interests, raising associated ethical concerns and bringing dark patterns into focus as a significant research topic. Manipulative design strategies are widely used in user interfaces (UI) primarily to guide user behavior in ways that favor service providers, often at the cost of the users themselves. This paper addresses three main challenges in dark pattern research: inconsistencies and incompleteness in classification, limitations of detection tools, and insufficient comprehensiveness in existing datasets. In this study, we propose a comprehensive analytical framework--the Dark Pattern Analysis Framework (DPAF). Using this framework, we developed a taxonomy comprising 68 types of dark patterns, each annotated in detail to illustrate its impact on users, potential scenarios, and real-world examples, validated through industry surveys. Furthermore, we evaluated the effectiveness of current detection tools and assessed the completeness of available datasets. Our findings indicate that, among the 8 detection tools studied, only 31 types of dark patterns are identifiable, resulting in a coverage rate of just 45.5%. Similarly, our analysis of four datasets, encompassing 5,561 instances, reveals coverage of only 30 types of dark patterns, with an overall coverage rate of 44%. Based on the available datasets, we standardized classifications and merged datasets to form a unified image dataset and a unified text dataset. These results highlight significant room for improvement in the field of dark pattern detection. This research not only deepens our understanding of dark pattern classification and detection tools but also offers valuable insights for future research and practice in this domain.

cs.HC↗

arXiv2024

SoK: Detection and Repair of Accessibility Issues

There is an increasing global emphasis on information accessibility, with numerous researchers actively developing automated tools to detect and repair accessibility issues, thereby ensuring that individuals with diverse abilities can independently access software products and services. However, current research still encounters significant challenges in two key areas: the absence of a comprehensive taxonomy of accessibility issue types, and the lack of comprehensive analysis of the capabilities of detection and repair tools, as well as the status of corresponding datasets. To address these challenges, this paper introduces the Accessibility Issue Analysis (AIA) framework. Utilizing this framework, we develop a comprehensive taxonomy that categorizes 55 types of accessibility issues across four pivotal dimensions: Perceivability, Operability, Understandability, and Robustness. This taxonomy has been rigorously recognized through a questionnaire survey (n=130). Building on this taxonomy, we conduct an in-depth analysis of existing detection and repair tools, as well as the status of corresponding datasets. In terms of tools, our findings indicate that 14 detection tools can identify 31 issue types, achieving a 56.3% rate (31/55). Meanwhile, 9 repair tools address just 13 issue types, with a 23.6% rate. In terms of datasets, those for detection tools cover 21 issue types, at a 38.1% coverage rate, whereas those for repair tools cover only 7 types, at a 12.7% coverage rate.

cs.SE↗

arXiv2024

Dual Mixed Volume

We define and study the dual mixed volume rational function of a sequence of polytopes, a dual version of the mixed volume polynomial. This concept has direct relations to the adjoint polynomials and the canonical forms of polytopes. We show that dual mixed volume is additive under mixed subdivisions, and is related by a change of variables to the dual volume of the Cayley polytope. We study dual mixed volume of zonotopes, generalized permutohedra, and associahedra. The latter reproduces the planar $\phi^3$-scalar amplitude at tree level.

math.CO↗