# Evaluating Software Process Models for Multi-Agent Class-Level Code Generation

arXiv: https://arxiv.org/abs/2511.09794 · HTML: https://mlsanity.com/paper/2511.09794/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 22 of its 52 references that are in the map.
- First posted on arXiv 12 Nov 2025.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 22 of 22)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [ChatDev: Communicative Agents for Software Development](https://mlsanity.com/paper/2307.07924/) (2023)
- [A Survey on Large Language Models for Code Generation](https://mlsanity.com/paper/2406.00515/) (2024)
- [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025)
- [BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions](https://mlsanity.com/paper/2406.15877/) (2024)
- [LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods](https://mlsanity.com/paper/2412.05579/) (2024)
- [Agentless: Demystifying LLM-based Software Engineering Agents](https://mlsanity.com/paper/2407.01489/) (2024)
- [CodeT5+: Open Code Large Language Models for Code Understanding and Generation](https://mlsanity.com/paper/2305.07922/) (2023)
- [ClassEval: A Manually-Crafted Benchmark for Evaluating LLMs on Class-level Code Generation](https://mlsanity.com/paper/2308.01861/) (2023)
- [A Survey on Code Generation with LLM-based Agents](https://mlsanity.com/paper/2508.00083/) (2025)
- [PanGu-Coder2: Boosting Large Language Models for Code with Ranking Feedback](https://mlsanity.com/paper/2307.14936/) (2023)
- [A Survey of Large Language Models for Code: Evolution, Benchmarking, and Future Trends](https://mlsanity.com/paper/2311.10372/) (2023)
- [Unifying the Perspectives of NLP and Software Engineering: A Survey on Language Models for Code](https://mlsanity.com/paper/2311.07989/) (2023)
- [Reasoning Runtime Behavior of a Program with LLM: How Far Are We?](https://mlsanity.com/paper/2403.16437/) (2024)
- [ClarifyGPT: Empowering LLM-based Code Generation with Intention Clarification](https://mlsanity.com/paper/2310.10996/) (2023)
- [SOEN-101: Code Generation by Emulating Software Process Models Using Large Language Model Agents](https://mlsanity.com/paper/2403.15852/) (2024)
- [CoCoST: Automatic Complex Code Generation with Online Searching and Correctness Testing](https://mlsanity.com/paper/2403.13583/) (2024)
- [DeepCode AI Fix: Fixing Security Vulnerabilities with Large Language Models](https://mlsanity.com/paper/2402.13291/) (2024)
- [Chain-of-Programming (CoP) : Empowering Large Language Models for Geospatial Code Generation](https://mlsanity.com/paper/2411.10753/) (2024)
- [Order Matters! An Empirical Study on Large Language Models' Input Order Bias in Software Fault Localization](https://mlsanity.com/paper/2412.18750/) (2024)
- [EvoGPT: Leveraging LLM-Driven Seed Diversity to Improve Search-Based Test Suite Generation](https://mlsanity.com/paper/2505.12424/) (2025)

## Built on by (most cited first, 2 of 2)

- [Architecture as Capability Equalizer for Coding Agents](https://mlsanity.com/paper/2608.21747/) (2026)
- [AgentRoom: Concurrent Multi-Agent Coding in a CRDT-Backed Shared Workspace](https://mlsanity.com/paper/2608.23740/) (2026)

## Abstract (arXiv)

Modern software systems require code that is not only functional but also maintainable and well-structured. Although Large Language Models (LLMs) are increasingly used to automate software development, most studies focus on isolated, single-agent function-level generation. This work examines how process structure and role specialization shape multi-agent LLM workflows for class-level code generation. We simulate a Waterfall-style development cycle covering Requirement, Design, Implementation, and Testing using three LLMs (GPT-4o-mini, DeepSeek-Chat, and Claude-3.5-Haiku) on 100 Python tasks from the ClassEval benchmark. Our findings show that multi-agent workflows reorganize, rather than consistently enhance, model performance. Waterfall-style collaboration produces cleaner and more maintainable code but often reduces functional correctness (-37.8% for GPT-4o-mini and -39.8% for DeepSeek-Chat), with Claude-3.5-Haiku as a notable exception (+9.5%). Importantly, process constraints shift failure characteristics: structural issues such as missing code decrease, while semantic and validation errors become more frequent. Among all stages, Testing exerts the strongest influence by improving verification coverage but also introducing new reasoning failures, whereas Requirement and Design have comparatively modest effects. Overall, this study provides empirical evidence that software process structure fundamentally alters how LLMs reason, collaborate, and fail, revealing inherent trade-offs between rigid workflow discipline and flexible problem-solving in multi-agent code generation.