Qing Huang

Application Programming Interfaces (APIs) are central to modern software development, yet the growing scale and semantic complexity of APIs present significant challenges for developers in learning and using them effectively. API knowledge graphs (KGs) offer a structured solution by organizing API entities and their relations, but existing construction methods face a trade-off: schema-guided methods ensure precision but suffer from limited richness, while schema-free methods capture richer semantics but often lack reliability. To address this challenge, we propose Explore-Construct-Verify ( ECV ), a three-stage framework for API KG construction using large language models (LLMs). The framework first explores candidate schema in a bottom-up manner, then constructs the KG with schema-guided extraction, and finally applies human-AI collaborative verification based on association rule mining. This design preserves LLMs’ ability to discover domain-specific knowledge while enabling efficient post-hoc validation. We conduct extensive experiments to evaluate the effectiveness of our method. Compared to the state-of-the-art method EDC, ECV improves KG construction F1 score by 31.3%. Component analysis shows that the exploration module improves KG richness (recall) by 267.9%, the fully connected schema strategy increases richness by 50.0%, and the verification module improves KG reliability (precision) by 47.1%. ECV maintains an average F1 around 0.69 both across six programming languages and across five LLMs, demonstrating strong cross-language and cross-model generalizability. We further validate the practical utility of the constructed KG through an API recommendation task, where KG-enhanced LLMs outperform LLM-only baselines by 26.3% in accuracy.

The success of prompt learning when adapted to the fully-qualified type name (FQN) inference has been demonstrated in the literature. However, the understanding of its success is limited in model outputs and model structures. In this paper, we conduct a thorough study on the behaviors of prompt learning in FQN inference from the perspectives of sensitivity and usefulness. Rather than simply masking some knowledge, we first perform sensitivity analysis on five aspects to reveal how much FQN knowledge to include, how much to mask, and where to mask, and then yield an efficient configuration strategy. We further conduct a usefulness analysis in three aspects to demonstrate the superiority of the proposed configuration strategy. This suggests that the strong performance of our model is attributable to the homogeneity among large code pre-training, FQN prompt learning, and type inference as a fill-in-blank task. Finally, we summarize a practical guideline on best practices and pitfalls to avoid when applying prompt learning to FQN inference and other software engineering tasks.

Network protocols are the foundation of modern communication, yet their implementations often contain semantic vulnerabilities stemming from inadequate understanding of specification semantics. Existing gray-box and black-box testing approaches lack semantic modeling of protocols, making it difficult to precisely express testing intent and cover boundary conditions. Moreover, they typically rely on coarse-grained oracles such as crashes, which are inadequate for identifying deep semantic vulnerabilities. To address these limitations, we present a semantics-aware fuzzing framework, SemFuzz. The framework leverages large language models to extract structured semantic rules from RFC documents and generates test cases that intentionally violate these rules to encode specific testing intents. It then detects deep semantic vulnerabilities by comparing the observed responses with the expected ones. Evaluation on seven widely deployed protocol implementations shows that SemFuzz identified sixteen potential vulnerabilities, ten of which have been confirmed. Among the confirmed vulnerabilities, five were previously unknown and four have been assigned CVEs. These results demonstrate the effectiveness of SemFuzz in detecting semantic vulnerabilities.

Textual Vulnerability Descriptions (TVDs) are crucial for security analysts to understand and address software vulnerabilities. However, the key aspect inconsistencies in TVDs from different repositories pose challenges for achieving a comprehensive understanding of vulnerabilities. Existing approaches aim to mitigate inconsistencies by aligning TVDs with external knowledge bases, but they often discard valuable information and fail to synthesize comprehensive representations. In this paper, we propose a domain-constrained LLM-based synthesis framework for unifying key aspects of TVDs. Our framework consists of three stages: 1) Extraction, guided by rule-based templates to ensure all critical details are captured; 2) Self-evaluation, using domain-specific anchor words to assess semantic variability across sources; and 3) Fusion, leveraging information entropy to reconcile inconsistencies and prioritize relevant details. This framework improves synthesis performance, increasing the F1 score for key aspect augmentation from 0.82 to 0.87, while enhancing comprehension and efficiency by over 30\%. We further develop Digest Labels, a practical tool for visualizing TVDs, which human evaluations show significantly boosts usability.

Snapshot compressive imaging (SCI) captures a 3D hyperspectral image (HSI) using a 2D compressive measurement and reconstructs the desired 3D HSI from that 2D measurement. The effective reconstruction method thus is crucial in SCI. Despite recent successes of deep learning (DL)-based methods over traditional approaches, they often ignore the intrinsic characteristics of HSI and are trained for a specific imaging system using sufficient paired datasets. To address this, we propose a novel self-supervised HSI reconstruction framework called low-rank tensor meets deep prior (LDMeet), which couples model-driven and data-driven methods. The design of LDMeet is inspired by the traditional model-driven low-rank tensor prior constructed based on domain knowledge, which can explore the intrinsic global spatial-spectral correlation of HSI and make the reconstruction method interpretable. To further utilize the powerful learning ability of DL-based approaches, we introduce a self-supervised spatial-spectral guided network (SSG-Net) into LDMeet to learn the implicit deep spatial-spectral prior of HSI without requiring training data, making it adaptable to various imaging systems. An efficient alternating direction method of multiplier (ADMM) is designed to solve the LDMeet model. Comprehensive experiments confirm that our LDMeet achieves superior results compared to self-supervised HSI reconstruction methods, while also yielding competitive results with supervised learning methods.

Smart contracts, which automatically execute transactions based on predefined conditions, play a crucial role in asset and money exchanges. Due to their involvement in significant financial transactions, these contracts are attractive targets for hackers, leading to substantial financial losses through exploitable vulnerabilities. While various program analysis methods such as Oyente, Mythril, and Securify have been proposed to address these security concerns, they rely on rule-based patterns that are time-consuming to develop and offer limited coverage. Deep learning methods present an alternative by automatically learning code features to detect vulnerabilities. However, existing approaches face critical challenges, including feature limitations and lack of interpretability. To address these gaps, we propose the interpretable smart contract vulnerability detector, a Graph Isomorphism Network (GIN)-based vulnerability prediction model for smart contracts, enhanced with code subgraph explanations. Our approach identifies and incorporates 43 domain-specific features, augmenting GIN with domain knowledge attention mechanisms to improve vulnerability prediction. In addition, we develop an interpreter called SubgraphV, which provides explanations for vulnerability predictions through interpreted subgraphs. Our model demonstrates superior performance over traditional tools, achieving F1 score improvements from 0.254 to 0.489 on a dataset of 103 smart contract function vulnerabilities. SubgraphV outperforms existing explainability methods like GNNexplainer, PGExplainer, and SubgraphX in pinpointing vulnerabilities, accurately reflecting vulnerability patterns, and enhancing the understanding of vulnerabilities.

Network clustering (graph clustering) plays a crucial role in discovering the inherent structures within networks. MapReduce-based structural clustering algorithm for networks (MRSCANs) designed based on MapReduce's parallel computing model, efficiently handles large-scale data. However, MRSCAN can only be tested through experiments, and its correctness cannot be guaranteed. To address this issue, this article has achieved the first implementation of functional MRSCAN modeling and subjected it to rigorous mechanized verification in Isabelle. First, based on Google's MapReduce model type definition and higher-order generic functions, a general MapReduce-based algorithm functional modeling framework is constructed across the fundamental global phases of Map, Shuffle, and Reduce. Moreover, diverse strategies are devised during the Shuffle phases according to user requirements, enhancing the applicability and generality of the MapReduce functional modeling framework. Second, formalizing the definition of MRSCAN, is delineated into four key steps: similarity calculation, core calculation, dimension expansion, and structural clustering. Furthermore, the MapReduce functional modeling framework is applied to these four steps to achieve the functional modeling of MRSCAN, which improves efficiency compared to other structural clustering algorithm for networks (SCANs) algorithms. Lastly, a verification framework for MapReduce-based algorithms is proposed at both the global and shuffle stages. Based on this framework, the correctness and reliability of MRSCAN are ensured. The model framework and verification framework of MapReduce-based algorithms proposed in this article can not only address functional modeling and verification of MRSCAN but also provide a reference for a series of other MapReduce-based functional program designs and proofs.

APIs have intricate relations that can be described in text and represented as knowledge graphs to aid software engineering tasks. Existing relation extraction methods have limitations, such as limited API text corpus, and are affected by the characteristics of the input text. To address these limitations, we propose utilizing large language models (LLMs) (e.g., GPT-3.5) as a neural knowledge base for API relation inference. This approach leverages the entire Web used to pre-train LLMs as a knowledge base and is insensitive to the context and complexity of input texts. To ensure accurate inference, we design an AI chain consisting of three AI modules: API Fully Qualified Name (FQN) Parser, API Knowledge Extractor, and API Relation Decider. The accuracy of the API FQN Parser and API Relation Decider is 0.81 and 0.83, respectively. Using the generative capacity of the LLM and our approach’s inference capability, we achieve an average F1 value of 0.76 under the three datasets, significantly higher than the state-of-the-art method’s average F1 value of 0.40. Compared to the original CoT and modularized CoT methods, our AI chain design has improved the performance of API relation inference by 71% and 49%, respectively. Meanwhile, the prompt ensembling strategy enhances the performance of our approach by 32%. The API relations inferred by our method can be further organized into structured forms to provide support for other software engineering tasks.