New AI framework HLS-Seek optimizes hardware design generation

Researchers have developed HLS-Seek, a novel framework for generating hardware designs from C/C++ code that prioritizes Quality of Results (QoR) such as latency and resource utilization. This system utilizes reinforcement learning with a comparative proxy reward model, which avoids the need for full synthesis in the loop and achieves high accuracy in predicting Pareto-optimal designs. HLS-Seek also incorporates an uncertainty-aware Monte Carlo dropout switching mechanism to refine its proxy model with real synthesis data, leading to improved performance and faster training compared to existing methods. AI

IMPACT This framework could accelerate hardware design by improving the efficiency and accuracy of AI-driven code generation for synthesis.

RANK_REASON Research paper detailing a new AI framework for hardware design. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• C Cpp Programming Languages
• GPT-5.1
• High-level synthesis
• HLS-Eval
• HLS-Seek
• Monte Carlo
• Proxy Comparative Reward Reinforcement Learning
• Qingyun Zou
• QoR-Aware Code Generation
• reinforcement learning
• Vitis HLS

• paper
• model release
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →