embedding_method.hpp File Reference

LLM-embedding-based code duplication detection. More...

#include <cstdio>
#include <string>
#include <vector>
#include <filesystem>
#include <arkanjo/methods/method.hpp>

Include dependency graph for embedding_method.hpp:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
class	EmbeddingMethod
	Semantic duplication detector using LLM embeddings. More...

Detailed Description

LLM-embedding-based code duplication detection.

Implements a duplication detector that computes semantic similarity between function bodies using Hugging Face embeddings (model: jinaai/jina-embeddings-v2-base-code), delegating the embedding and pairwise comparison to a Python helper script located in the third-party folder.

Definition in file embedding_method.hpp.