~ similar to 2610.02376· 20 results
The paper proposes the Intelligent Computing Architecture Model (ICAM), a six-layer framework that unifies disparate concepts in model-native computing by viewing the LLM stack through a dual-plane ar…
This paper presents a design methodology for Compound AI systems that can approach the accuracy of monolithic models while reducing latency and cost.
The paper introduces HORIZON, a self-evolving agent framework for hardware design using git operations.
This tutorial explores advances and challenges in deploying large language model-based agentic systems across industries, with a focus on reasoning and planning, multi-agent coordination, and evaluati…
Chenyu Wang, Zishen Wan, Jeffrey Ma, Shvetank Prakash +7 more
This paper introduces ArchEval, a benchmark and platform for evaluating LLM agents on computer architecture design and optimization.
Boxiu Li, Zimo Wen, Yijia Fan, Junxiang Lei +22 more
Argus is a persistent, self-evolving runtime system that achieves better long-horizon reasoning results than Direct Copilot in seven GPT-5.5 benchmark arenas, with improvements in solve-input tokens,…
The paper argues that current 'on-the-fly' AI agent design lacks necessary software engineering rigor and proposes an 'AI Workflow Store' to provide hardened, reusable, and reliable agent workflows.
Chenyu Wang, Jiahe Caroline Shi, David Kong, Duane S. Boning +3 more
AgentDSE is a simulator-in-the-loop methodology that uses a large language model to automate architectural design space exploration.
Paras Dahal, Anton Bakhtin, Taco Cohen, Zhengxing Chen +8 more
The paper introduces agentic meta-reasoning, a structured reasoning process for controlling the execution of complex tasks.
Aditya Kumar, Zhihan Lei, Jerry Yan, Joshua W. Momo +5 more
The paper proposes a modular agent framework and novel learning methods to design and optimize practical, cost-effective, and controllable LLM-based agentic systems.
Yuebo Luo, Ahmad Sedigh Baroughi, Philip Stachura, Le Chen +3 more
HLSmith is a framework that uses expert-guided techniques to translate C/C++ programs into optimized FPGA accelerators using HLS.
Yangshen Deng, Jiaqi Ruan, Yeqi Huang, Nick Lau +2 more
The paper introduces InfraDev, a benchmark for evaluating AI infrastructure development, and shows that existing benchmarks are inadequate for assessing real-world tasks.
Peter Kulits, Yiqing Xu, R. Kenny Jones, Cordelia Schmid +1 more
A benchmark for evaluating text-conditioned LEGO-set design agents that can produce physically buildable designs.
Jixuan Chen, Jianzhi Shen, Haoqiang Kang, Zhi Hong +9 more
This paper introduces AgentSpec, a modular specification framework for embodied agents that enables the isolation and comparison of component contributions to agent behavior.
Ruiyin Li, Yiran Zhang, Xiyu Zhou, Yangxiao Cai +5 more
The paper introduces MAAD, a multi-agent framework that autonomously transforms software requirements into comprehensive, multi-view architectural blueprints, significantly improving completeness and…