arXiv subjects

Zhijie Wang

Publications and source records attributed to Zhijie Wang.

At least 19 recordsLinked to original sources

arXiv2026

Rethinking Code Performance Benchmarks for LLMs

Many function-level performance benchmarks have been proposed to evaluate whether large language models (LLMs) can generate efficient programs. However, results on these benchmarks often show that LLM-generated implementations have little or no execution-time difference from canonical solutions. In this paper, we revisit four popular benchmarks: EffiBench, Enamel, EvalPerf, and Mercury. We evaluate 1,538 tasks under more rigorous setting by running each task 30 times and assessing the runtime differences between the canonical solutions and benchmark-provided performant implementations with statistical testing. With the benchmark-provided test suites, only 6.11% of the performant implementations are significantly faster than the canonical solutions. In a manual analysis of 308 non-significant tasks, 99 performant implementations contain no meaningful performance change, while 209 contain potential performance improvements that are not exposed by the original tests. These results suggest that the main limitation is not only the evaluation method, but also the limited sufficiency of the benchmark-provided performance tests. To address this limitation, we propose an LLM-based multi-agent framework to generate performance-oriented tests that expose runtime differences more effectively than the original tests. The framework uses three separate agents to generate, diagnose, and repair deterministic tests that preserve functional correctness while better exposing performance differences. Across 1,345 benchmark tasks for which the original tests found no significant performance difference, tests generated by our framework with DeepSeek-v3.1 and GPT-4o reveal statistically significant improvements in 24.01% and 25.43% of the tasks, respectively, outperforming the SOTA LLM-based performance test generation method.

cs.SE↗

arXiv2026

LogSemFuse: Semantic Evidence Fusion for Explainable Log Anomaly Detection

Log anomaly detection is critical for reliability monitoring and failure diagnosis in modern software systems. Existing model-based detectors provide useful anomaly signals, but they can still miss anomalous sessions and typically expose only scores or labels rather than the operational semantics behind a decision. This lack of semantic evidence limits their ability to explain why a session is anomalous, even when the final anomaly label is correct. The gap matters in practice because operators need to distinguish urgent failures from benign deviations and trace suspicious sessions back to concrete operational behavior. LLMs can recover richer log semantics, but using them as standalone detectors or repeatedly generating free-form explanations can be costly and difficult to reuse. We present LogSemFuse, an evidence-guided plug-in framework that enhances existing backbone detectors without replacing their original pipelines. LogSemFuse combines backbone predictions with reusable semantic evidence from local event patterns, LLM-based semantic reasoning, and cluster-derived executable rules to produce both anomaly decisions and evidence-based explanations. The resulting output reports the final label together with the semantic evidence that supports it, such as fired local patterns, triggered rules, and LLM rationale. We evaluate LogSemFuse on HDFS, BGL, and Liberty using DeepLog, LogAnomaly, LogBERT, and NeuralLog as backbones. LogSemFuse improves every non-perfect baseline, preserves the already perfect case, recovers 98.8% of backbone false negatives, and produces explanations preferred over direct LLM explanations in a human study. These gains require only modest and stable inference-time overhead, showing that semantic augmentation can improve detection effectiveness and interpretability without imposing large runtime costs.

cs.SE↗

arXiv2026

Understanding Software Defect Prediction: A Large-scale Empirical Study Across Uncertainty Quantification and Performance Evaluation

Software defect prediction (SDP) classifiers produce probabilities used for inspection prioritization, threshold tuning, and risk communication. Probability-based uncertainty quantification (UQ) characterizes prediction confidence, but whether common UQ metrics reliably indicate performance and calibration remains unclear. We conducted a large-scale empirical study of probability-based UQ for SDP. We evaluated five UQ metrics, six performance metrics, and three calibration metrics for 16 representative classifiers. We analyzed these relationships under two prediction settings: within-project defect prediction (WPDP), using 36 benchmark datasets, and cross-project defect prediction (CPDP), using 32 feature-compatible datasets. Results showed that UQ was highly context-dependent. Under WPDP, UQ correlated more consistently with false positive rate and AUC than with MCC, F1 score, and other metrics; these correlations also varied across classifier categories and dataset collections. Performance and calibration were related but not interchangeable; classifiers with strong discrimination could still exhibit large calibration error. Under CPDP, several UQ-performance and UQ-calibration correlations weakened or reversed, indicating that uncertainty signals do not reliably transfer across projects. Thus, UQ should be evaluated against specific performance objectives. Calibration should be assessed independently using multiple metrics. Transferred probabilities should be revalidated before guiding quality-assurance decisions.

cs.SE↗

arXiv2026

OpenRath: Session-Centered Runtime State for Agent Systems

Modern agent systems often suffer from fragmented runtime state: transcripts, tool effects, memory events, workspace placement, branch provenance, and replay evidence are recorded separately and become difficult to inspect or reproduce. OpenRath addresses this issue with a PyTorch-like programming model for multi-agent, multi-session systems. The analogy concerns the role of a central first-class runtime abstraction, not tensor computation. Its core abstraction is Session, the runtime value passed between agents and workflows. A Session is branchable, inspectable, replayable, backend-aware, and composable. It records conversation chunks, sandbox placement, lineage metadata, token usage, pending work, and tool evidence, while defining where memory interactions enter the runtime record. Since this state is carried by the same value used in program execution, fork, merge, and replay become explicit runtime operations rather than states reconstructed from external traces. OpenRath further defines Sandbox, Tool, Agent, Memory, Workflow, and Selector, with Selector turning control flow into runtime-routed decisions. This report presents the programming model, architecture, audited milestones, and evidence protocol. Its claims are limited to controlled runtime properties, while broad quantitative comparisons, live-provider quality, optional-backend availability, and memory quality are left for follow-on evaluation. The central thesis is that Session provides agent systems with a first-class runtime value for auditable composition.

cs.SE↗

arXiv2026

vLLM Semantic Router: Signal Driven Decision Routing for Mixture-of-Modality Models

As large language models (LLMs) diversify across modalities, capabilities, and cost profiles, the problem of intelligent request routing: selecting the right model for each query at inference time, has become a critical systems challenge. We present vLLM Semantic Router, a signal-driven decision routing framework for Mixture-of-Modality (MoM) model deployments. The architecture follows two complementary Shannon-inspired views. In the information-theoretic regime, signal extraction reduces the entropy of "which model?" by distilling routing-relevant information from raw queries. In the Boolean-algebraic regime, the decision engine composes functionally complete routing policies from signal conditions. The central innovation is composable signal orchestration: thirteen heterogeneous signal types, spanning sub-millisecond heuristics and neural classifiers for semantics, safety, and modality, are composed through configurable Boolean decision rules into deployment-specific routing policies, so that fundamentally different scenarios (multi-cloud enterprise, privacy-regulated, cost-optimized) are expressed as different configurations over the same architecture. Matched decisions drive semantic model routing via thirteen selection algorithms, while per-decision plugin chains enforce safety constraints including a three-stage HaluGate hallucination detection pipeline and a lightweight episodic memory system with ReflectionGate for personalized multi-turn context. A typed neural-symbolic DSL specifies these routing policies and compiles them to multiple deployment targets, enabling configuration-first adaptation without code changes. Together, these components show that composable signal orchestration enables a single framework to serve diverse deployment scenarios with differentiated cost, privacy, and safety policies.

cs.NI↗

arXiv2026

FALAT: Tracing Failures in LLM Agent Trajectories via Dependency-Guided Search

LLM-based agents increasingly solve complex tasks through long trajectories involving reasoning steps, tool calls, and inter-agent communication. However, when these agents fail, it is often unclear which agent caused the failure and which step introduced the decisive error. This attribution problem is challenging because mistakes can propagate across the trajectory: later actions may appear incorrect, but only because they depend on an earlier corrupted state. Therefore, failure attribution cannot be treated as independent step-level classification. We propose FALAT, a diagnostic framework for failure attribution in LLM agent trajectories. FALAT frames attribution as a dependency-guided search problem. It first constructs an expectation of how the task should be solved and uses this expectation to identify suspicious regions in the trajectory. It then traces dependencies among decisions, tool outputs, and agent messages to distinguish error-introducing steps from steps that merely inherit or propagate prior mistakes. Finally, FALAT evaluates whether correcting a candidate step would be sufficient to recover the expected outcome, allowing it to identify both the responsible agent and the decisive failure step. We evaluate FALAT on the Who&When benchmark, which includes both algorithm-generated and hand-crafted multi-agent failure trajectories. The results show that FALAT consistently improves responsible-agent and decisive-step attribution. Its best configurations achieve 46.0% step-level accuracy on algorithm-generated trajectories and 29.1% on the more challenging hand-crafted trajectories, outperforming specialized attribution baselines and direct prompting with standalone LLMs. These findings suggest that dependency-aware reasoning is essential for reliable failure diagnosis in LLM agent systems.

cs.AI↗

arXiv2026

Understanding the Fundamental Design Decisions of Retrieval-Augmented Generation Systems

Retrieval-Augmented Generation (RAG) has emerged as a critical technique for enhancing large language model (LLM) capabilities. However, practitioners face significant challenges when making RAG deployment decisions. While existing research prioritizes algorithmic innovations, a systematic gap persists in understanding fundamental engineering trade-offs that determine RAG success. We present the first comprehensive study of three universal RAG deployment decisions: whether to deploy RAG, how much information to retrieve, and how to integrate retrieved knowledge effectively. Through systematic experiments across three LLMs and six datasets spanning question answering and code generation tasks, we reveal critical insights: (1) RAG deployment must be highly selective, with variable recall thresholds and failure modes affecting up to 12.6\% of samples even with perfect documents. (2) Optimal retrieval volume exhibits task-dependent behavior QA tasks show universal patterns (5-10 documents optimal) while code generation requires scenario-specific optimization. (3) Knowledge integration effectiveness depends on task and model characteristics, with code generation benefiting significantly from prompting methods while question answering shows minimal improvement. These findings demonstrate that universal RAG strategies prove inadequate. Effective RAG systems require context-aware design decisions based on task characteristics and model capabilities. Our analysis provides evidence-based guidance for practitioners and establishes foundational insights for principled RAG deployment. Our code, data and artifacts are publicly available at https://github.com/ShengmingZ/RAG_Benchmark_Code_QA.

cs.SE↗

arXiv2026

SSDAU: Structured Semantic Data Augmentation for Joint Entity and Relation Extraction

Joint Entity and Relation Extraction (JERE) is highly sensitive to training data quality, making data augmentation a natural way to improve generalization. However, existing augmentation methods often weaken entity relevance and disrupt semantic structure, limiting their effectiveness for JERE. In this paper, we propose \textbf{Structured Semantic Data Augmentation (SSDAU)}, a method designed to preserve triple-aware semantic structure during augmentation. SSDAU segments text by entity labels, captures semantic features through context-aware encoding, and restructures entity semantics to generate augmented data. To distinguish semantically similar entities, SSDAU combines contextualized embeddings with traditional similarity scores. To reduce topic inconsistency, we apply BERTopic-based filtering to remove irrelevant augmentations. We evaluate SSDAU on datasets with different annotation types and compare its performance on five representative JERE models against seven popular augmentation baselines. Experiments show that SSDAU generates semantically consistent data, is more robust to ambiguity than non-LLM methods (8.95\% vs. 23.58\% average relative F1 decrease), and significantly outperforms strong alternatives in most settings.

cs.CL↗

arXiv2026

Learning First Integrals via Backward-Generated Data and Guided Reinforcement Learning

The discovery of first integrals is of fundamental scientific importance for understanding conservation laws in dynamical systems. However, existing symbolic computation tools and Large Language Models (LLMs) remain limited on this task because high-quality training data are scarce and successful solutions often depend on mathematical intuition. This paper presents FISolver, an LLM-based solver developed to address this challenge. First, we introduce a "Backward Generation" algorithm that systematically builds large-scale datasets of (differential equation, first integral) pairs by deriving differential equations from sampled integrals, thereby alleviating the data scarcity bottleneck. Second, we apply supervised fine-tuning to a compact mathematical model and further improve its performance through reinforcement learning with a Levenshtein Distance-based shaped reward. In addition, we design data synthesis and blending strategies that support effective adaptation to difficult problem families from sparse examples. Experiments show that FISolver, while requiring substantially lower computational cost, significantly outperforms larger mathematical LLMs and commercial solvers such as Mathematica on challenging benchmarks, indicating a new data-driven route for automated discovery of first integrals.

cs.LG↗

arXiv2026

Beam Test Characterization of Silicon Microstrip Detector Flight-Model Ladders for the AMS-02 Upgrade

The AMS-02 experiment plans to install a new silicon microstrip tracker layer (Layer-0) on top of the existing detector, increasing the cosmic-ray acceptance by a factor of 3. Layer-0 employs a design in which multiple silicon microstrip detectors (SSDs) are connected in series to form long detector ladders. We present a detailed performance study of the flight-model ladders using a 350~GeV mixed hadron beam at the CERN SPS. The study focuses on the following aspects: (i) the performance of ladders with different numbers of SSDs, for which the intrinsic spatial resolution at normal incidence varies from $9.5~μ\mathrm{m}$ to $11.4~μ\mathrm{m}$ for ladders composed of 8 to 12 SSDs; (ii) the response consistency for particles impacting on the \emph{Head} and \emph{Tail} regions of the ladder; and (iii) the dependence of the detector performance on the particle incidence angle.

physics.ins-det↗

arXiv2026

GRPO and Reflection Reward for Mathematical Reasoning in Large Language Models

The enhancement of reasoning capabilities in large language models (LLMs) has garnered significant attention, with supervised fine-tuning (SFT) and reinforcement learning emerging as dominant paradigms. While recent studies recognize the importance of reflection in reasoning processes, existing methodologies seldom address proactive reflection encouragement during training. This study focuses on mathematical reasoning by proposing a four-stage framework integrating Group Relative Policy Optimization (GRPO) with reflection reward mechanisms to strengthen LLMs' self-reflective capabilities. Besides, this approach incorporates established accuracy and format reward. Experimental results demonstrate GRPO's state-of-the-art performance through reflection-encouraged training, with ablation studies confirming the reflection reward's pivotal role. Comparative evaluations demonstrate full-parameter SFT's superiority over low-rank adaptation (LoRA) despite heightened computational demands. Building on these cumulative findings, this research substantiates GRPO's methodological significance in post-training optimization and envisions its potential to serve as a pivotal enabler for future LLM-based intelligent agents through the synergistic integration of cognitive rewards with dynamic environmental interactions.

cs.AI↗

arXiv2025

Matrix Completion Via Reweighted Logarithmic Norm Minimization

Low-rank matrix completion (LRMC) has demonstrated remarkable success in a wide range of applications. To address the NP-hard nature of the rank minimization problem, the nuclear norm is commonly used as a convex and computationally tractable surrogate for the rank function. However, this approach often yields suboptimal solutions due to the excessive shrinkage of singular values. In this letter, we propose a novel reweighted logarithmic norm as a more effective nonconvex surrogate, which provides a closer approximation than many existing alternatives. We efficiently solve the resulting optimization problem by employing the alternating direction method of multipliers (ADMM). Experimental results on image inpainting demonstrate that the proposed method achieves superior performance compared to state-of-the-art LRMC approaches, both in terms of visual quality and quantitative metrics.

cs.CV↗

arXiv2025

Class-agnostic 3D Segmentation by Granularity-Consistent Automatic 2D Mask Tracking

3D instance segmentation is an important task for real-world applications. To avoid costly manual annotations, existing methods have explored generating pseudo labels by transferring 2D masks from foundation models to 3D. However, this approach is often suboptimal since the video frames are processed independently. This causes inconsistent segmentation granularity and conflicting 3D pseudo labels, which degrades the accuracy of final segmentation. To address this, we introduce a Granularity-Consistent automatic 2D Mask Tracking approach that maintains temporal correspondences across frames, eliminating conflicting pseudo labels. Combined with a three-stage curriculum learning framework, our approach progressively trains from fragmented single-view data to unified multi-view annotations, ultimately globally coherent full-scene supervision. This structured learning pipeline enables the model to progressively expose to pseudo-labels of increasing consistency. Thus, we can robustly distill a consistent 3D representation from initially fragmented and contradictory 2D priors. Experimental results demonstrated that our method effectively generated consistent and accurate 3D segmentations. Furthermore, the proposed method achieved state-of-the-art results on standard benchmarks and open-vocabulary ability.

cs.CV↗

arXiv2025

Twist dominates bending in the liquid crystal organization of bacteriophage DNA

DNA frequently adopts liquid-crystalline conformations in both cells and viruses. The Oseen--Frank framework provides a powerful continuum description of these phases through three elastic moduli: splay ($K_1$), twist or cholesteric ($K_2$), and bending ($K_3$). While $K_1$ is typically assumed to dominate, the relative magnitude of $K_2$ and $K_3$ in confined DNA remains poorly understood. Here, we combine cryo-electron microscopy, liquid-crystal modeling, and knot theory to quantify this relationship in bacteriophage P4, whose genome is partially organized in a spool-like liquid-crystalline phase. We first show experimentally that the ordered DNA occupies three concentric layers within the capsid. We then formulate an Oseen--Frank model for this geometry and use it, together with the measured layer radii, to estimate the elastic ratio $α= K_3/K_2$. We find $α\approx 0.0064$, indicating that twist elasticity overwhelmingly dominates bending. To validate this result, we perform Langevin dynamics simulations of DNA trajectories and classify the resulting knots. The predicted knot distribution agrees with experimental data from P4, demonstrating consistency between elasticity, topology, and observed genome organization.

cond-mat.soft↗

arXiv2025

Knotted DNA Configurations in Bacteriophage Capsids: A Liquid Crystal Theory Approach

Bacteriophages, viruses that infect bacteria, store their micron long DNA inside an icosahedral capsid with a typical diameter of 40 nm to 100 nm. Consistent with experimental observations, such confinement conditions induce an arrangement of DNA that corresponds to a hexagonal chromonic liquid-crystalline phase, and increase the topological complexity of the genome in the form of knots. A mathematical model that implements a chromonic liquid-crystalline phase and that captures the changes in topology has been lacking. We adopt a mathematical model that represents the viral DNA as a pair of a vector field and a line. The vector field is a minimizer of the total Oseen-Frank energy for nematic liquid crystals under chromonic constraints, while the line is identified with the tangent to the field at selected locations, representing the central axis of the DNA molecule. The fact that the Oseen-Frank functional assigns infinite energy to topological defects (point defects in two dimensions and line defects in three dimensions) precludes the presence of singularities and, in particular, of knot structures. To address this issue, we begin with the optimal vector field and helical line, and propose a new algorithm to introduce knots through stochastic perturbations associated with splay and twist deformations, modeled by means of a Langevin system. We conclude by comparing knot distributions generated by the model and by interpreting them in the context of previously published experimental results. Altogether, this work relies on the synergy of modeling, analysis and computation in the study of viral DNA organization in capsids.

cond-mat.soft↗

arXiv2025

GCC-Spam: Spam Detection via GAN, Contrastive Learning, and Character Similarity Networks

The exponential growth of spam text on the Internet necessitates robust detection mechanisms to mitigate risks such as information leakage and social instability. This work addresses two principal challenges: adversarial strategies employed by spammers and the scarcity of labeled data. We propose a novel spam-text detection framework GCC-Spam, which integrates three core innovations. First, a character similarity network captures orthographic and phonetic features to counter character-obfuscation attacks and furthermore produces sentence embeddings for downstream classification. Second, contrastive learning enhances discriminability by optimizing the latent-space distance between spam and normal texts. Third, a Generative Adversarial Network (GAN) generates realistic pseudo-spam samples to alleviate data scarcity while improving model robustness and classification accuracy. Extensive experiments on real-world datasets demonstrate that our model outperforms baseline approaches, achieving higher detection rates with significantly fewer labeled examples.

cs.LG↗

arXiv2025

Fibring structures of ideals in Roe algebras and their $K$-theories

In this paper, we investigate the ideal structure of Roe algebras for metric spaces beyond the scope of Yu's property A. Using the tool of rank distributions, we establish fibring structures for the lattice of ideals in Roe algebras and draw the border of each fibre by introducing the so-called ghostly ideals together with geometric ideals. We also provide coarse geometric criteria to ensure the coincidence of geometric and ghostly ideals and calculate their $K$-theories, which can be helpful to analyse obstructions to the coarse Baum-Connes conjecture on the level of ideals.

math.OA↗

arXiv2025

VLATest: Testing and Evaluating Vision-Language-Action Models for Robotic Manipulation

The rapid advancement of generative AI and multi-modal foundation models has shown significant potential in advancing robotic manipulation. Vision-language-action (VLA) models, in particular, have emerged as a promising approach for visuomotor control by leveraging large-scale vision-language data and robot demonstrations. However, current VLA models are typically evaluated using a limited set of hand-crafted scenes, leaving their general performance and robustness in diverse scenarios largely unexplored. To address this gap, we present VLATest, a fuzzing framework designed to generate robotic manipulation scenes for testing VLA models. Based on VLATest, we conducted an empirical study to assess the performance of seven representative VLA models. Our study results revealed that current VLA models lack the robustness necessary for practical deployment. Additionally, we investigated the impact of various factors, including the number of confounding objects, lighting conditions, camera poses, unseen objects, and task instruction mutations, on the VLA model's performance. Our findings highlight the limitations of existing VLA models, emphasizing the need for further research to develop reliable and trustworthy VLA applications.

cs.SE↗