CodeBERTScore: an automatic metric for code generation, based on BERTScore

EVIL (Exploiting software VIa natural Language) is an approach to automatically generate software exploits in assembly/Python language from descriptions in natural language. The approach leverages Neural Machine Translation (NMT) techniques and a dataset that we developed for this work.

Neural search engine for discovering semantically similar Python repositories on GitHub

Augmenting the Interpretability of GraphCodeBERT for Code Similarity Tasks

Repository about small code models

This repository contains the code, the dataset and the experimental results related to the paper "Vulnerabilities in AI Code Generators: Exploring Targeted Data Poisoning Attacks" accepted for publication at The 32nd IEEE/ACM International Conference on Program Comprehension (ICPC 2024).

🕵️‍♂️ ML project to identify malicious web payloads, aimed at boosting the effectiveness of WAFs and IDSs.

Improving Source Code Similarity Detection with GraphCodeBERT and Additional Feature Integration

Code of our paper "Method-Level Bug Severity Prediction using Source Code Metrics and LLMs" which is accepted to ISSRE 2023.

Advanced Detection of Source Code Clones via an Ensemble of Unsupervised Similarity Measures

Fine-tuning CodeBERT for Vulnerability Detection

Fine-tuning CodeBERT with AST-based Vectors for Code Translation

This repository contains experiments on comparing the similarity of Python repositories using ML models.

Fine-tuned CodeBERT model detecting real security vulnerabilities (CWE-119, CWE-416, CWE-476+) in C/C++ — trained on 190K+ CVE functions, deployed as a FastAPI server, GitHub Action, and VS Code extension.

Performs Code Summarization, Bug Detection, Bug Removal using different Natural language processing models including Garph CodeBERT, GREAT, GNN, CoText etc.

Implementation and dataset for A Zero-Shot Framework for Cross-Project Vulnerability Detection in Source Code (Empirical Software Engineering, 2026).

CodeBERT + LoRA fine-tuning for C/C++ vulnerability detection | F1 = 74.3% | PyTorch, HuggingFace Transformers, PEFT

CodeOpt: A framework for optimizing code performance using Two-Stage Sampling, Few-Shot Learning, and Iterative Self-Reflection with support for Genetic Algorithm Inspired Chain-of-Thought (GA-COT).

The study uses the IRSE/FIRE dataset and explores the impact of combining original C code data with Python-derived silver-standard

A project for determining the similarity of python repositories based on embedding approach