Advances in AI-Assisted Software Development

The field of software development is witnessing significant advancements with the integration of Artificial Intelligence (AI) and Large Language Models (LLMs). Recent studies have demonstrated the potential of AI-assisted tools in improving code review, automated unit test generation, and bug detection. One notable trend is the development of LLM-based approaches for automated program repair, which has shown promising results in generating correct and efficient fixes. Additionally, researchers have explored the use of multi-agent systems and finite state machines to improve the accuracy and effectiveness of AI-assisted software development tools. The application of AI and LLMs in software development has also led to improved code quality, reduced debugging time, and enhanced overall productivity. Noteworthy papers in this area include studies on AI-assisted code review, LLM-based unit test generation, and bug detection using multi-agent systems. Notable papers include: AI-Assisted Fixes to Code Review Comments at Scale, which presents a system for providing AI-assisted fixes for code review comments. Input Reduction Enhanced LLM-based Program Repair, which proposes an approach to reduce test inputs for LLM-based program repair. BugScope, which introduces a multi-agent system for bug detection that emulates human auditors.

Sources

AI-Assisted Fixes to Code Review Comments at Scale

Impact of Code Context and Prompting Strategies on Automated Unit Test Generation with Modern General-Purpose Large Language Models

Input Reduction Enhanced LLM-based Program Repair

BugScope: Learn to Find Bugs Like Human

GasAgent: A Multi-Agent Framework for Automated Gas Optimization in Smart Contracts

Do AI models help produce verified bug fixes?

StaAgent: An Agentic Framework for Testing Static Analyzers

From Contracts to Code: Automating Smart Contract Generation with Multi-Level Finite State Machines

Seed&Steer: Guiding Large Language Models with Compilable Prefix and Branch Signals for Unit Test Generation

AssertFlip: Reproducing Bugs via Inversion of LLM-Generated Passing Tests

An Empirical Study on Embodied Artificial Intelligence Robot (EAIR) Software Bugs

YATE: The Role of Test Repair in LLM-Based Unit Test Generation

Built with on top of