mlsanityData to 7 Oct 2026

LIFT: LLM-Based Pragma Insertion for HLS via GNN Supervised Fine-Tuning

arXiv 2504.21187 ↗First posted 29 Apr 2025cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

FPGAs are increasingly adopted in datacenter environments for their reconfigurability and energy efficiency. High-Level Synthesis (HLS) tools have eased FPGA programming by raising the abstraction level from RTL to untimed C/C++, yet attaining high performance still demands expert knowledge and iterative manual insertion of optimization pragmas to modify the microarchitecture. To address this challenge, we propose LIFT, a large language model (LLM)-based coding assistant for HLS that automatically generates performance-critical pragmas given a C/C++ design. We fine-tune the LLM by tightly integrating and supervising the training process with a graph neural network (GNN), combining the sequential modeling capabilities of LLMs with the structural and semantic understanding of GNNs necessary for reasoning over code and its control/data dependencies. On average, LIFT produces designs that improve performance by 3.52x and 2.16x than prior state-of the art AutoDSE and HARP respectively, and 66x than GPT-4o.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 18 of its 36 references that are in the map. First posted on arXiv 29 Apr 2025.

Builds on

18 of 36 references.

References · 18

  1. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,099
  2. 2023VerilogEval: Evaluating Large Language Models for Verilog Code Generationcited by 246
  3. 2025KernelBench: Can LLMs Write Efficient GPU Kernels?cited by 206
  4. 2023VeriGen: A Large Language Model for Verilog Code Generationcited by 197
  5. 2020Enabling Language Models to Fill in the Blankscited by 121
  6. 2024AnalogCoder: Analog Circuit Design via Training-Free Code Generationcited by 82
  7. 2024Harnessing Diversity for Important Data Selection in Pretraining Large Language Modelscited by 34
  8. 2021IronMan: GNN-assisted Design Space Exploration in High-Level Synthesis via Reinforcement Learningcited by 26
  9. 2020AutoDSE: Enabling Software Programmers to Design Efficient FPGA Acceleratorscited by 25
  10. 2022High-Level Synthesis Performance Prediction using GNNs: Benchmarking, Modeling, and Advancingcited by 25
  11. 2024HLSPilot: LLM-based High-Level Synthesiscited by 24
  12. 2024Automated C/C++ Program Repair for High-Level Synthesis via Large Language Modelscited by 18
6 more
  1. 2024Cross-Modality Program Representation Learning for Electronic Design Automation with High-Level Synthesiscited by 10
  2. 2021DB4HLS: A Database of High-Level Synthesis Design Space Explorationscited by 8
  3. 2022Democratizing Domain-Specific Computingcited by 6
  4. 2024Automatic Hardware Pragma Insertion in High-Level Synthesis: A Non-Linear Programming Approachcited by 6
  5. 2024Efficient Task Transfer for HLS DSEcited by 4
  6. 2024Accelerating Large Language Model Pretraining via LFR Pedagogy: Learn, Focus, and Reviewcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 7.

8 papers.

  1. 2025iDSE: Navigating Design Space Exploration in High-Level Synthesis Using LLMscited by 6
  2. 2026Bench4HLS: End-to-End Evaluation of LLMs in High-Level Synthesis Code Generationcited by 2
  3. 2026Agent Factories for High Level Synthesis: How Far Can General-Purpose Coding Agents Go in Hardware Optimization?cited by 1
  4. 2026Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verificationcited by 1
  5. 2026A3D: Agentic AI flow for autonomous Accelerator Designcited by 0
  6. 2026HLSmith: An Expert-Guided Agentic Framework for C/C++-to-HLS Translationcited by 0
  7. 2026Can Agents Design Better Chips with a Higher Level Abstraction?cited by 0
  8. 2026SyntheticHLS: Building Diverse Synthetic High-Level Synthesis Datasets using LLMscited by 0