# AgentDSE: Reasoning-Augmented Architectural Design Space Exploration

arXiv: https://arxiv.org/abs/2606.21836 · HTML: https://mlsanity.com/paper/2606.21836/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 19 of its 28 references that are in the map.
- First posted on arXiv 20 Jun 2026.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 19 of 19)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [ChipNeMo: Domain-Adapted LLMs for Chip Design](https://mlsanity.com/paper/2311.00176/) (2023)
- [ChatEDA: A Large Language Model Powered Autonomous Agent for EDA](https://mlsanity.com/paper/2308.10204/) (2023)
- [Gemmini: Enabling Systematic Deep-Learning Architecture Evaluation via Full-Stack Integration](https://mlsanity.com/paper/1911.09925/) (2019)
- [AI Agentic Programming: A Survey of Techniques, Challenges, and Opportunities](https://mlsanity.com/paper/2508.11126/) (2025)
- [The Championship Simulator: Architectural Simulation for Education and Competition](https://mlsanity.com/paper/2210.14324/) (2022)
- [ORFS-agent: Tool-Using Agents for Chip Design Optimization](https://mlsanity.com/paper/2506.08332/) (2025)
- [LLM-DSE: Searching Accelerator Parameters with LLM Agents](https://mlsanity.com/paper/2505.12188/) (2025)
- [SCALE-Sim v3: A modular cycle-accurate systolic accelerator simulator for end-to-end system analysis](https://mlsanity.com/paper/2504.15377/) (2025)
- [ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design](https://mlsanity.com/paper/2306.08888/) (2023)
- [ArchAgent: Agentic AI-driven Computer Architecture Discovery](https://mlsanity.com/paper/2602.22425/) (2026)
- [Computer Architecture's AlphaZero Moment: Automated Discovery in an Encircled World](https://mlsanity.com/paper/2604.03312/) (2026)
- [Economy of Minds: Emerging Multi-Agent Intelligence with Economic Interactions](https://mlsanity.com/paper/2606.02859/) (2026)
- [DOSA: Differentiable Model-Based One-Loop Search for DNN Accelerators](https://mlsanity.com/paper/2509.10702/) (2025)
- [EPIM: Efficient Processing-In-Memory Accelerators based on Epitome](https://mlsanity.com/paper/2311.07620/) (2023)
- [gem5 Co-Pilot: AI Assistant Agent for Architectural Design Space Exploration](https://mlsanity.com/paper/2510.19577/) (2025)
- [QuArch: A Benchmark for Evaluating LLM Reasoning in Computer Architecture](https://mlsanity.com/paper/2510.22087/) (2025)
- [GenAI for Systems: Recurring Challenges and Design Principles from Software to Silicon](https://mlsanity.com/paper/2602.15241/) (2026)

## Abstract (arXiv)

Traditional architectural design space exploration (DSE) is highly inefficient, typically requiring tens of thousands of simulator evaluations across various optimization methods. This inefficiency arises because conventional methods treat the simulator as a black-box oracle. In contrast, human architects effectively guide exploration by reasoning through physical constraints, performance bottlenecks, data reuse, and workload structures. To bridge this gap, we introduce AgentDSE, a simulator-in-the-loop methodology driven by a general-purpose large language model (LLM) coding agent. AgentDSE automates this architectural-reasoning loop without requiring model fine-tuning, precomputed design databases, or domain-specific optimizer code. Across deep neural network (DNN) accelerator mapping, hardware/software co-design, and CPU cache-hierarchy optimization, AgentDSE achieves competitive or better design quality with up to two orders of magnitude fewer evaluations. AgentDSE also produces inspectable traces that surface architectural hypotheses, performance cliffs, implicit priors, and simulator artifacts, making every search decision traceable rather than buried in optimizer state.