ARCS: Agentic Retrieval-Augmented Code Synthesis with Iterative Refinement
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908613474254848 |
|---|---|
| author | Bhattarai, Manish Cordova, Miguel Vu, Minh Santos, Javier Boureima, Ismael O'Malley, Dan |
| author_facet | Bhattarai, Manish Cordova, Miguel Vu, Minh Santos, Javier Boureima, Ismael O'Malley, Dan |
| contents | We present Agentic Retrieval-Augmented Code Synthesis (ARCS), a system that improves LLM-based code generation without fine-tuning. ARCS operates through a budgeted synthesize-execute-repair loop over a frozen model: it retrieves relevant code context before generation, proposes candidates, executes them against tests, and repairs based on execution feedback. This retrieval-before-generation design reduces hallucination and accelerates convergence. We formalize ARCS as a state-action process with provable guarantees on termination, monotonic improvement, and bounded cost. A tiered controller (Small/Medium/Large) trades latency for accuracy predictably. On HumanEval, ARCS achieves up to 87.2% pass@1 with Llama-3.1-405B, surpassing CodeAgent (82.3%) while using simpler control than tree-search methods. On TransCoder, it achieves >= 90% accuracy on most translation pairs. On a LANL scientific corpus, it improves CodeBLEU by +0.115 over baseline RAG. ARCS provides a practical, reproducible approach to reliable code synthesis using existing LLM checkpoints. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_20434 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ARCS: Agentic Retrieval-Augmented Code Synthesis with Iterative Refinement Bhattarai, Manish Cordova, Miguel Vu, Minh Santos, Javier Boureima, Ismael O'Malley, Dan Software Engineering Artificial Intelligence We present Agentic Retrieval-Augmented Code Synthesis (ARCS), a system that improves LLM-based code generation without fine-tuning. ARCS operates through a budgeted synthesize-execute-repair loop over a frozen model: it retrieves relevant code context before generation, proposes candidates, executes them against tests, and repairs based on execution feedback. This retrieval-before-generation design reduces hallucination and accelerates convergence. We formalize ARCS as a state-action process with provable guarantees on termination, monotonic improvement, and bounded cost. A tiered controller (Small/Medium/Large) trades latency for accuracy predictably. On HumanEval, ARCS achieves up to 87.2% pass@1 with Llama-3.1-405B, surpassing CodeAgent (82.3%) while using simpler control than tree-search methods. On TransCoder, it achieves >= 90% accuracy on most translation pairs. On a LANL scientific corpus, it improves CodeBLEU by +0.115 over baseline RAG. ARCS provides a practical, reproducible approach to reliable code synthesis using existing LLM checkpoints. |
| title | ARCS: Agentic Retrieval-Augmented Code Synthesis with Iterative Refinement |
| topic | Software Engineering Artificial Intelligence |
| url | https://arxiv.org/abs/2504.20434 |