# Automatically Benchmarking LLM Code Agents through Agent-Driven Annotation and Evaluation

arXiv: https://arxiv.org/abs/2510.24358 · HTML: https://mlsanity.com/paper/2510.24358/

## Key facts

- Cited by 9 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 21 of its 55 references that are in the map.
- First posted on arXiv 28 Oct 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, 21 of 21)

- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Prometheus: Inducing Fine-grained Evaluation Capability in Language Models](https://mlsanity.com/paper/2310.08491/) (2023)
- [BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions](https://mlsanity.com/paper/2406.15877/) (2024)
- [GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models](https://mlsanity.com/paper/2508.06471/) (2025)
- [MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark](https://mlsanity.com/paper/2402.04788/) (2024)
- [JudgeLM: Fine-tuned Large Language Models are Scalable Judges](https://mlsanity.com/paper/2310.17631/) (2023)
- [PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization](https://mlsanity.com/paper/2306.05087/) (2023)
- [Generative Judge for Evaluating Alignment](https://mlsanity.com/paper/2310.05470/) (2023)
- [An Empirical Study of LLM-as-a-Judge for LLM Evaluation: Fine-tuned Judge Model is not a General Substitute for GPT-4](https://mlsanity.com/paper/2403.02839/) (2024)
- [HumanEval-XL: A Multilingual Code Generation Benchmark for Cross-lingual Natural Language Generalization](https://mlsanity.com/paper/2402.16694/) (2024)
- [CodeUltraFeedback: An LLM-as-a-Judge Dataset for Aligning Large Language Models to Coding Preferences](https://mlsanity.com/paper/2403.09032/) (2024)
- [LLM-based Agents Suffer from Hallucinations: A Survey of Taxonomy, Methods, and Directions](https://mlsanity.com/paper/2509.18970/) (2025)
- [SciReplicate-Bench: Benchmarking LLMs in Agent-driven Algorithmic Reproduction from Research Papers](https://mlsanity.com/paper/2504.00255/) (2025)
- [Automated Benchmark Generation for Repository-Level Coding Tasks](https://mlsanity.com/paper/2503.07701/) (2025)
- [ProjectEval: A Benchmark for Programming Agents Automated Evaluation on Project-Level Code Generation](https://mlsanity.com/paper/2503.07010/) (2025)
- [On the Impacts of Contexts on Repository-Level Code Generation](https://mlsanity.com/paper/2406.11927/) (2024)
- [ExecRepoBench: Multi-level Executable Code Completion Evaluation](https://mlsanity.com/paper/2412.11990/) (2024)
- [Codev-Bench: How Do LLMs Understand Developer-Centric Code Completion?](https://mlsanity.com/paper/2410.01353/) (2024)
- [Reassessing Extractive QA Datasets at Scale: LLM-as-a-Judge and In-Depth Analyses](https://mlsanity.com/paper/2504.11972/) (2025)
- [RepoMasterEval: Evaluating Code Completion via Real-World Repositories](https://mlsanity.com/paper/2408.03519/) (2024)
- [Deep-Bench: Deep Learning Benchmark Dataset for Code Generation](https://mlsanity.com/paper/2502.18726/) (2025)

## Built on by (most cited first, 9 of 9)

- [Scaling Embeddings Outperforms Scaling Experts in Language Models](https://mlsanity.com/paper/2601.21204/) (2026)
- [ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders](https://mlsanity.com/paper/2607.21217/) (2026)
- [SaaSBench: Exploring the Boundaries of Coding Agents in Long-Horizon Enterprise SaaS Engineering](https://mlsanity.com/paper/2605.17526/) (2026)
- [Asuka-Bench: Benchmarking Code Agents on Underspecified User Intent and Multi-Round Refinement](https://mlsanity.com/paper/2606.05920/) (2026)
- [SWE-AGI: Benchmarking Specification-Driven Software Construction with MoonBit in the Era of Autonomous Agents](https://mlsanity.com/paper/2602.09447/) (2026)
- [Compiling Large Multi-Modal Requirement Documents into Runnable Software Systems: From an Agentic Test-Driven Perspective](https://mlsanity.com/paper/2602.13723/) (2026)
- [From Skills to Talent: Organising Heterogeneous Agents as a Real-World Company](https://mlsanity.com/paper/2604.22446/) (2026)
- [Embodied-BenchClaw: An Autonomous Multi-Agent System for Embodied Spatial Intelligence Benchmark Construction](https://mlsanity.com/paper/2606.11909/) (2026)
- [AutoDataBench: Can Agents Write the Data That Feeds the Self-Improvement Loop?](https://mlsanity.com/paper/2609.35025/) (2026)

## Abstract (arXiv)

Recent advances in code agents have enabled automated software development at the project level, supported by large language models (LLMs). However, existing benchmarks for code agent evaluation face two major limitations. First, creating high-quality project-level evaluation datasets requires extensive domain expertise, leading to prohibitive annotation costs and limited diversity. Second, while recent Agent-as-a-Judge paradigms address the rigidity of traditional unit tests by enabling flexible metrics, their reliance on In-Context Learning (ICL) with general LLMs often results in inaccurate assessments that misalign with human standards. To address these challenges, we propose an agent-driven benchmark construction pipeline that leverages human supervision to efficiently generate diverse project-level tasks. Based on this, we introduce PRDBench, comprising 50 real-world Python projects across 20 domains, each with structured Product Requirement Documents (PRDs) and comprehensive criteria. Furthermore, to overcome the inaccuracy of general LLM judges, we propose a highly reliable evaluation framework powered by a specialized, fine-tuned model. Based on Qwen3-Coder-30B, our dedicated PRDJudge achieves over 90% human alignment in fixed-interface scenarios. Extensive experiments demonstrate that our suite provides a scalable, robust, and highly accurate framework for assessing state-of-the-art code agents.