5 papers
Agentic Refactoring: An Empirical Study of AI Coding Agents
Kosei Horikawa, Hao Li, Yutaro Kashiwa +3
Agentic coding tools, such as OpenAI Codex, Claude Code, and Cursor, are transforming the software engineering landscape. These AI-powered systems function as autonomous teammates…
Understanding Prompt Management in GitHub Repositories: A Call for Best Practices
Hao Li, Hicham Masri, Filipe R. Cogo +3
The rapid adoption of foundation models (e.g., large language models) has given rise to promptware, i.e., software built using natural language prompts. Effective management of pro…
On the Effect of Token Merging on Pre-trained Models for Code
Mootez Saad, Hao Li, Tushar Sharma +1
Tokenization is a fundamental component of language models for code. It involves breaking down the input into units that are later passed to the language model stack to learn high-…
Can We Recycle Our Old Models? An Empirical Evaluation of Model Selection Mechanisms for AIOps Solutions
Yingzhe Lyu, Hao Li, Heng Li +1
AIOps (Artificial Intelligence for IT Operations) solutions leverage the tremendous amount of data produced during the operation of large-scale systems and machine learning models…
Towards Refining Developer Questions using LLM-Based Named Entity Recognition for Developer Chatroom Conversations
Pouya Fathollahzadeh, Mariam El Mezouar, Hao Li +2
In software engineering chatrooms, communication is often hindered by imprecise questions that cannot be answered. Recognizing key entities can be essential for improving question…