Review Credibility Assessment
Building robust multimodal models that detect both human-written and AI-generated fake reviews using textual, visual, temporal, and relational cues.
About
Multimodal learning, trustworthy NLP, and code understanding.
I work on multimodal learning, trustworthy NLP, and code understanding, with current projects spanning fake review detection and multimodal large models for software engineering.
My recent work includes MDCFN for robust multimodal review credibility assessment and CodeOCR, a study of vision-language models for code understanding.
I am currently an M.Sc. student in Library and Information Studies at Hohai University, and I also work with the LLM for Software Engineering Lab at Shanghai Jiao Tong University.
My recent research focuses on robust multimodal review credibility assessment and the effectiveness of vision-language models for code understanding. I am particularly interested in combining strong empirical evaluation with practical engineering systems.
News
Full CV- 2026-04-21 ClassEval-Pro was accepted to AIWare 2026.
- 2026-04-17 CodeOCR was accepted to ISSTA 2026.
- 2024 Joined the LLM for Software Engineering Lab at Shanghai Jiao Tong University as a research assistant.
- 2024 Started the M.Sc. program in Library and Information Studies at Hohai University.
- 2024 Won the Excellent Work Award at the Intel Mini Hackathon for a fine-tuned LLM project.
Research Interests
Current research interests
My work sits at the intersection of multimodal machine learning, information credibility, and software engineering.
Multimodal Code Understanding
Exploring whether rendered code images and visual cues can let multimodal large models match or exceed text-based baselines in software engineering tasks.
Applied Data Mining
Using machine learning and information retrieval methods to study practical information systems problems with an emphasis on reliable evaluation and strong engineering execution.
Publications
Selected publications
Selected publications and ongoing work.
CodeOCR: On the Effectiveness of Vision Language Models in Code Understanding
Proceedings of ISSTA 2026 · 2026
Studies code-as-image representations for multimodal code understanding and shows how visual encoding can improve efficiency while remaining competitive on downstream tasks.
Conference
ClassEval-Pro: A Cross-Domain Benchmark for Class-Level Code Generation
ACM AIWare 2026 Data and Benchmark Track Submission · 2026
Introduces ClassEval-Pro, a benchmark of 300 class-level code generation tasks across 11 domains, built through an automated three-stage pipeline with complexity enhancement, cross-domain class composition, and real-world GitHub code integration. Each task is validated by an LLM Judge Ensemble and test suites with over 90% line coverage. Experiments on five frontier LLMs under five generation strategies show that the best model reaches only 45.6% class-level Pass@1, while error analysis highlights logic and dependency errors as the main bottlenecks.
Conference
Multi-Detector Credibility Fusion Network: A Neural Architecture for Robust Multimodal Review Credibility Assessment
International Journal of Intelligent Systems · 2026
Presents MDCFN, a multimodal architecture for robust review credibility assessment across textual, visual, and relational signals.
Under Review
Experience
Research and professional experience
Research-first, with industry experience that informs implementation and systems thinking.
Research
Research Assistant
LLM for Software Engineering Lab (LLMSE), Shanghai Jiao Tong University
Advisor: Prof. Xiaodong Gu
- Led engineering for the full CodeOCR experimental pipeline, a systematic study of multimodal LLMs for code understanding.
- Implemented visual context compression that renders code into images and achieved up to 8x token compression while preserving semantic integrity.
- Ran evaluations on code completion and clone detection, showing that visual cues such as syntax highlighting can match or outperform text-only baselines in specific settings.
Independent Researcher
Institute of Management Science, Hohai University
- Proposed the Multi-Detector Credibility Fusion Network for detecting sophisticated fake reviews, including human-written and AI-generated content.
- Designed a hierarchical fusion mechanism combining textual-temporal, visual, and relational graph branches for complementary credibility cues.
- Built a large annotated multimodal dataset with 33k+ reviews and 50k+ images, including LLM-generated content, and achieved 98.77% accuracy beyond prior baselines.
Education & Industry
M.Sc. in Library and Information Studies
Hohai University
- Focus on data mining with GPA 87/100.
- Relevant coursework: Business Intelligence Analysis and Mining, Advanced Information Retrieval, Machine Learning Applications.
B.Sc. in Information Management and Information System
Hohai University
- Graduated with GPA 85/100.
- Relevant coursework: Statistics, Database Principles, Data Structures, Information Security, Calculus, Linear Algebra.
Software Engineer (Python Backend)
Inspur Morning Cloud Technologies Co., Ltd.
- Developed core modules for a Human Capital Management system serving enterprise clients.
- Improved complex import and export workflows by implementing tree-traversal logic for multi-level headers, reducing processing time substantially.
- Received the R&D Rising Star Award for technical innovation and secured one pending patent.
Projects
Selected projects
A few implementation-heavy projects that reflect both experimentation and execution.
Simulating Conversations with Fine-Tuned LLMs
- Fine-tuned Qwen-7B-Chat with LoRA on a custom Rednote dataset to mimic persona-based conversational styles.
- Deployed the model for real-time interaction and won the Excellent Work Award at the Intel Mini Hackathon.
E-sports Player Style Clustering Analysis
- Built a pipeline to crawl more than 100 matches of top-ranked players and extract 11 key performance indicators.
- Applied K-means clustering to categorize playstyles and generate strategic insights for team composition.