Sujoy Roychowdhury

Code Comprehensor: Hierarchical Documentation and Cross-Modal Retrieval for Large Repositories

This work proposes Code Comprehensor, a hybrid retrieval-reasoning framework combining CodeGraph, a code dependency callgraph, CodeScribe, which uses code-aware LLMs to generate hierarchical documentation across repository call graphs, and CodeContrast, a CLIP-style dual-encoder model trained on CodeScribe output to le...