{"ID":2870648,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.13557","arxiv_id":"2509.13557","title":"MACO: A Multi-Agent LLM-Based Hardware/Software Co-Design Framework for CGRAs","abstract":"Coarse-Grained Reconfigurable Arrays (CGRAs) offer high performance and energy efficiency across domains, yet design remains difficult due to a vast, interdependent space and costly manual iteration. We present MACO, an open-source multi-agent LLM framework for CGRA HW/SW co-design. MACO generates and refines architectures through four stages: HW/SW Co-design, Design Error Correction, Best-Design Selection, and Evaluation \u0026 Feedback, iteratively improves power, performance, and area (PPA) via agent reasoning and closed-loop feedback. To traverse the space efficiently, we introduce exponentially decaying exploration; to accelerate convergence, we incorporate an LLM self-learning mechanism that adaptively selects promising candidate CGRAs. In addition, we propose a rule-based mechanism to correct CGRA design errors. Evaluated against other state-of-the-art methods, MACO achieves superior PPA while substantially reducing human effort, highlighting the promise of LLM-driven automation for practical CGRA design.","short_abstract":"Coarse-Grained Reconfigurable Arrays (CGRAs) offer high performance and energy efficiency across domains, yet design remains difficult due to a vast, interdependent space and costly manual iteration. We present MACO, an open-source multi-agent LLM framework for CGRA HW/SW co-design. MACO generates and refines architect...","url_abs":"https://arxiv.org/abs/2509.13557","url_pdf":"https://arxiv.org/pdf/2509.13557v6","authors":"[\"Zesong Jiang\",\"Yuqi Sun\",\"Qing Zhong\",\"Mahathi Krishna\",\"Deepak Patil\",\"Cheng Tan\",\"Jeff Zhang\"]","published":"2025-09-16T21:52:04Z","proceeding":"cs.AR","tasks":"[\"cs.AR\"]","methods":"[\"Large Language Model\",\"LoRA\"]","has_code":false}