Projects

The primary objective is to design, fine-tune and deploy a secure, internally-facing Large Language Model for auditors (ALLM) at Henan Provincial Audit Department. Built on Qwen2.5-VL, ALLM need deliver intelligent support for audit Q&A, regulatory compliance and table-data analysis. In this project, I worked with Lutong Zhang and Haonan Zhang to finish:

• Data Collection: Aggregating amounts of audit documentation, historical audit reports, government financial regulations (national and provincial level), accounting standards (e.g., Chinese GAAP), policy directives, compliance manuals, and best practice guidelines.
• Data Processing: Implementing data processing pipelines for heterogeneous audit sources, including: text extraction that applies PaddleOCR with regex-based digit correction for scanned documents, pdfplumber for layout-aware PDF, python-docx for DOCX files; spaCy + FinBERT for entity normalization of regulations/financial terms; DeepSeek-V2 classification for document categorization and UIE model for sensitivity tagging; LayoutLMv3 for context-aware semantic segmentation to preserve audit logic.
• QA Generation and Expert Annotation: Using Qwen-72B-Chat to produce question-answer pairs reflecting common audit scenarios, enriching the fine-tuning dataset; collaborating with senior auditors to annotate real audit cases, identifying key entities, risks, compliance issues, and generating relevant queries/responses.

This project aims to collect and quantify stomatal phenotypes from live field-grown peanut plants. By integrating phenotypic, genotypic and environmental data, we can perform Genome-Wide Association Study (GWAS) and Genotype-by-Environment Interaction Analysis (G×E) to identify key and highly stable regulatory genes. The ultimate goal is to discover superior genetic resources for peanut breeding under variable environments and enhance its productivity and adaptability. I worked with Quanling Zhao and Prof. Xiaohui Yang, as well as Dr. Chenyang Du and Prof. Chen Miao from State Key Laboratory of Crop Stress Adaptation and Improvement, to finish:

• GxEStat, a user-friendly R GUI package designed for G×E analysis. It integrates: (1) significance analysis based on the mixed effect model to determine whether GxE interactions significantly affect traits; (2) single-genotype and multi-genotype stability analysis to further studies the interactive effects as well as the relative superiority or inferiority of genotypes across environments.
• StomaD², an end-to-end stomatal phenotyping system based on diffusion-based restoration detection network. StomaD² supports real-time imaging and quantification of stomatal phenotypes (e.g., density, conductance) in field conditions. It is also compatible with both monocot and dicot crops, and accepts both destructive and non-destructive microscopy images.
• We propose a triphasic Hellinger Distance-based Intersection over Union (HDIoU) for oriented bounding box regression. HDIoU utilizes a dynamic indicator to determine the current optimization focus during training, allowing the loss function to continuously transition across three objective phases. This triphasic design facilitates more accurate and efficient localization of target contours, yielding precise phenotypic information for stomatal studies. Furthermore, HDIoU is inherently scale-invariant, making it particularly effective for detecting small objects such as stomata in high-resolution imagery.