mlsanityData to 13 Oct 2026

HLS-Seek: QoR-Aware Code Generation for High-Level Synthesis via Proxy Comparative Reward Reinforcement Learning

arXiv 2605.13536 ↗First posted 13 May 2026cs.LG cs.AIcited by 1

On the map: LLM alignment / RLHF / preference optimisation → Process reward models for reasoning

Abstract (arXiv)

High-Level Synthesis (HLS) compiles algorithmic C/C++ descriptions into hardware, with Quality of Results (QoR)---latency and resource utilization---critically governed by pragma configurations and code structure. Existing natural-language-to-HLS (NL-to-HLS) training approaches prioritize functional correctness while largely ignoring QoR. We observe that reinforcement learning (RL) for HLS does not require absolute synthesis results---only relative comparisons between candidates. Based on this insight, we propose HLS-Seek, a QoR-aware NL-to-HLS framework that avoids full synthesis-in-the-loop RL via a comparative proxy reward model achieving 99.53% Pareto-dominance accuracy. To prevent reward hacking, we introduce textituncertainty-aware Monte Carlo (MC) dropout switching that selectively invokes real Vitis HLS synthesis for low-confidence candidates and online updates the proxy, creating a self-improving reward system. HLS-Seek achieves 84.7% syntax correctness pass@1 and 81.4% functional correctness pass@5 on HLS-Eval~citeabikaram2025hlseval with only 7B parameters, surpassing GPT-5.1 on functional pass@5, while achieving 8.5× faster training than real-reward RL. On QoR evaluation, HLS-Seek achieves the lowest latency on 19/30 kernels and Pareto-dominates HLS-specific baselines on 9 kernels.

Key facts. Cited by 1 paper in the map (data through 13 Oct 2026). Placed in Process reward models for reasoning, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 17 of its 32 references that are in the map. First posted on arXiv 13 May 2026.

Builds on

17 of 32 references.

References · 17

  1. 2022Training language models to follow instructions with human feedbackcited by 15,856
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,091
  3. 2021Evaluating Large Language Models Trained on Codecited by 8,205
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,430
  5. 2024Qwen2.5-Coder Technical Reportcited by 1,179
  6. 2022UniXcoder: Unified Cross-Modal Pre-training for Code Representationcited by 360
  7. 2018Fast inference of deep neural networks in FPGAs for particle physicscited by 133
  8. 2024FlightLLM: Efficient Large Language Model Inference with a Complete Mapping Flow on FPGAscited by 64
  9. 2024C2HLSC: Leveraging Large Language Models to Bridge the Software-to-Hardware Design Gapcited by 16
  10. 2024C2HLSC: Can LLMs Bridge the Software-to-Hardware Design Gap?cited by 12
  11. 2025Exploring Code Language Models for Automated HLS-based Hardware Generation: Benchmark, Infrastructure and Analysiscited by 10
  12. 2025HLS-Eval: A Benchmark and Framework for Evaluating LLMs on High-Level Synthesis Design Taskscited by 10
5 more
  1. 2025SAGE-HLS: Syntax-Aware AST-Guided LLM for High-Level Synthesis Code Generationcited by 9
  2. 2024SynthAI: A Multi Agent Generative AI Framework for Automated Modular HLS Design Generationcited by 6
  3. 2025HLStrans: Dataset for C-to-HLS Hardware Code Synthesiscited by 3
  4. 2023LightRW: FPGA Accelerated Graph Dynamic Random Walkscited by 2
  5. 2023Data Processing with FPGAs on Modern Architecturescited by 1

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026HLSmith: An Expert-Guided Agentic Framework for C/C++-to-HLS Translationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026CodeScaler: Scaling Code LLM Training and Test-Time Inference via Reward Models
  2. 2024Optimizing High-Level Synthesis Designs with Retrieval-Augmented Large Language Models
  3. 2025Solver-Informed RL: Grounding Large Language Models for Authentic Optimization Modeling
  4. 2025VERIRL: Boosting the LLM-based Verilog Code Generation via Reinforcement Learning
  5. 2026Reinforcement Learning for Code Optimization
  6. 2026Improving HPC Code Generation Capability of LLMs via Online Reinforcement Learning with Real-Machine Benchmark Rewards
  7. 2025QeRL: Beyond Efficiency -- Quantization-enhanced Reinforcement Learning for LLMs
  8. 2025The Best of N Worlds: Aligning Reinforcement Learning with Best-of-N Sampling via max@k Optimisation