7 papers
COMPAS: Difficulty-Aware Joint Search for Optimizing Code Generation
Jingzhi Gong, Jie M. Zhang, Gunel Jahangirova +3
Code generation systems make each LLM call with a model, a prompt, and decoding settings. However, existing optimization methods usually tune only part of these choices or use one…
Agent4cs: A Multi-agent System for Code Summarization in Large Hierarchical Codebases
Yongjian Tang, Ezgi Sarikayak, Doruk Tuncel +2
Understanding large, complex codebases, especially those with obfuscated structures and incomplete documentation, remains a significant challenge. Existing code summarization solut…
An Empirical Study of Downstream Adaptation for Agent Skills
Xinjian Wu, Jingzhi Gong, Gunel Jahangirova +2
As Large Language Model (LLM) agents become integral to modern software systems, ``skills'' have emerged as a novel unit of software reuse, enabling developers to package workflows…
How Does Chunking Affect Retrieval-Augmented Code Completion? A Controlled Empirical Study
Xinjian Wu, Jingzhi Gong, Gunel Jahangirova +1
Retrieval-augmented generation (RAG) pipelines for code completion rely on chunking to segment source files into retrievable units, yet chunking strategies are typically adopted wi…
Benchmarking and Evaluating VLMs for Software Architecture Diagram Understanding
Shuyin Ouyang, Jie M. Zhang, Jingzhi Gong +7
Software architecture diagrams are important design artifacts for communicating system structure, behavior, and data organization throughout the software development lifecycle. Alt…
LLMs are Bug Replicators: An Empirical Study on LLMs' Capability in Completing Bug-prone Code
Liwei Guo, Sixiang Ye, Zeyu Sun +6
Large Language Models (LLMs) have demonstrated remarkable performance in code completion. However, the training data used to develop these models often contain a significant amount…