孙海龙｜北京航空航天大学｜学者主页

2026

基于深度学习的软件工程:进展、挑战与机遇

陈湘萍，胡星，黄袁，江贺，计卫星，姜艳杰，蒋炎岩，刘博，刘辉，李晓晨，连小利，孟国柱，彭鑫，孙海龙，石琳，王博，王翀，王加益，王甜甜，玄跻峰，夏鑫，杨已彪，杨艺欣，张莉，周毓明，张路

《中国科学:信息科学》

2026

Vulnerability identification by harnessing inter-connected multi-source information

Liyou Chen, Hailong Sun, Xiang Gao, Lin Shi, Yixin Yang, Yi Xu

Journal of Systems and Software

精选作品

期刊论文

基于深度学习的软件工程:进展、挑战与机遇

Vulnerability identification by harnessing inter-connected multi-source information

DepUpgrade: Automating Dependency Upgrade through State-Path Exploration

Generating Project-Specific Test Cases with Requirement Validation Intention

Type-aware LLM-based Test Generation for Python Programs

Generalizing Test Cases for Comprehensive Test Scenario Coverage

Efficient Bug Detection by Inferring Implicit API Contract of Pointer State Transition

Fusing LLMs and Genetic Algorithm for High-Quality Unit Test Generation

Attribution Analysis-based Concept Alignment: A Human-in-the-loop Data Debugging Framework

Beyond Superficial Tests: Adversarial Refinement for Reliable Property-Based Testing

D 3 em: A Dual-Layer Dynamic Debiasing Evaluation Mechanism for Client Contribution in Federated Learning

DeMuS: Learning Decoupled Matching and Scoring for Batch Zero-Shot Industrial Anomaly Detection

Reference-Based Retrieval-Augmented Unit Test Generation

Fed- : A tri-level federated fairness evaluation framework with statistical guarantees and privacy preservation

Enhanced Vulnerability Localization: Harmonizing Task-Specific Tuning and General LLM Prompting

Code Property Graph Meets Typestate: A Scalable Framework to Behavioral Bug Detection

Enhancing Automated Vulnerability Repair Through Dependency Embedding and Pattern Store

Software Engineering for OpenHarmony: A Research Roadmap

Deep learning-based software engineering: progress, challenges, and opportunities

API Misuse Detection via Probabilistic Graphical Model

ModelFoundry: A Tool for DNN Modularization and On-Demand Model Reuse Inspired by the Wisdom of Software Engineering

Investigating White-Box Attacks for On-Device Models

Reducing False Positives of Static Bug Detectors Through Code Representation Learning

Investigating and Detecting Silent Bugs in PyTorch Programs

Reusing Convolutional Neural Network Models through Modularization and Composition

Detecting Condition-Related Bugs with Control Flow Graph Neural Network

An Approach to Assessing the Health of Opensource Software Ecosystems

An error consistency based approach to answer aggregation in open-ended crowdsourcing

Improving fault localization and program repair with deep semantic features and transferred knowledge

Incorporating pixel proximity into answer aggregation for crowdsourced image segmentation

DreamLoc: A Deep Relevance Matching-Based Framework for bug Localization

Incorporating Multiple Features to Predict Bug Fixing Time with Neural Networks

Model learning: a survey of foundations, tools and applications

Teaching Active Human Learners

A Low-Code Development Framework for Constructing Industrial Apps

Data Poisoning Attacks on Crowdsourcing Learning

Learning to handle exceptions

Robust Adversarial Active Learning with a Novel Diversity Constraint

Find truth in the hands of the few: acquiring specific knowledge with crowdsourcing

Structured Probabilistic End-to-End Learning from Crowds

Retrieval-based neural source code summarization

Predicting Crowdsourcing Worker Performance with Knowledge Tracing

Combining Model Learning and Model Checking to Analyze Java Libraries

How are distributed bugs diagnosed and fixed through system logs?

Developer recommendation for Topcoder through a meta-learning based policy model

CONAN: A framework for detecting and handling collusion in crowdsourcing

A Novel Neural Source Code Representation Based on Abstract Syntax Tree

An Online Developer Profiling Tool Based on Analysis of GitLab Repositories

Testing-based Model Learning Approach for Legacy Components

大数据驱动的开发者社区中知识交流网络的分析

Identifying Potential Experts on Stack Overflow

Automatically Generating API Usage Patterns from Natural Language Queries

Service4Crowd

Profiling Developer Expertise across Software Communities with Heterogeneous Information Network Analysis

A Hybrid Approach to Developer Recommendation Based on Multi-relationship

On the Cost Complexity of Crowdsourcing

Entity-level sentiment analysis of issue comments

Semantically enhanced tag recommendation for software CQAs via deep learning

Context-aware result inference in crowdsourcing

Collusion-Proof Result Inference in Crowdsourcing

Budgeted Task Scheduling for Crowdsourced Knowledge Acquisition

A hybrid approach to code reviewer recommendation with collaborative filtering

Competition-aware task routing for contest based crowdsourced software development

Recommending crowdsourced software developers in consideration of skill improvement

A Decision Tree Based Quality Control Framework for Multi-phase Tasks in Crowdsourcing

Improving the Quality of Crowdsourced Image Labeling via Label Similarity

Achieving convergent causal consistency and high availability for cloud storage

Intelligent Development Environment and Software Knowledge Graph

Adaptive Result Inference for Collecting Quantitative Data With Crowdsourcing

Handling multi-dimensional complex queries in key-value data stores

An efficient and highly available framework of data recency enhancement for eventually consistent data stores

Adaptive trade‐off between consistency and performance in data replication

Don't get caught in the cold, warm-up your JVM: understand and eliminate JVM warm-up overhead in data-parallel systems

Incorporating external knowledge into crowd intelligence for more specific knowledge acquisition

Achieving convergent causal consistency and high availability with asynchronous replication

Effective Result Inference for Context-Sensitive Tasks in Crowdsourcing

Combining Crowd Contributions with Machine Learning to Detect Malicious Mobile Apps

Efficient Testing of Web Services with Mobile Crowdsourcing