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

arXiv: https://arxiv.org/abs/2504.21187 · HTML: https://mlsanity.com/paper/2504.21187/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 36 references that are in the map.
- First posted on arXiv 29 Apr 2025.

## Builds on (most cited first, 18 of 18)

- [Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning](https://mlsanity.com/paper/1801.07606/) (2018)
- [VerilogEval: Evaluating Large Language Models for Verilog Code Generation](https://mlsanity.com/paper/2309.07544/) (2023)
- [KernelBench: Can LLMs Write Efficient GPU Kernels?](https://mlsanity.com/paper/2502.10517/) (2025)
- [VeriGen: A Large Language Model for Verilog Code Generation](https://mlsanity.com/paper/2308.00708/) (2023)
- [Enabling Language Models to Fill in the Blanks](https://mlsanity.com/paper/2005.05339/) (2020)
- [AnalogCoder: Analog Circuit Design via Training-Free Code Generation](https://mlsanity.com/paper/2405.14918/) (2024)
- [Harnessing Diversity for Important Data Selection in Pretraining Large Language Models](https://mlsanity.com/paper/2409.16986/) (2024)
- [IronMan: GNN-assisted Design Space Exploration in High-Level Synthesis via Reinforcement Learning](https://mlsanity.com/paper/2102.08138/) (2021)
- [High-Level Synthesis Performance Prediction using GNNs: Benchmarking, Modeling, and Advancing](https://mlsanity.com/paper/2201.06848/) (2022)
- [AutoDSE: Enabling Software Programmers to Design Efficient FPGA Accelerators](https://mlsanity.com/paper/2009.14381/) (2020)
- [HLSPilot: LLM-based High-Level Synthesis](https://mlsanity.com/paper/2408.06810/) (2024)
- [Automated C/C++ Program Repair for High-Level Synthesis via Large Language Models](https://mlsanity.com/paper/2407.03889/) (2024)
- [Cross-Modality Program Representation Learning for Electronic Design Automation with High-Level Synthesis](https://mlsanity.com/paper/2406.09606/) (2024)
- [DB4HLS: A Database of High-Level Synthesis Design Space Explorations](https://mlsanity.com/paper/2101.00587/) (2021)
- [Democratizing Domain-Specific Computing](https://mlsanity.com/paper/2209.02951/) (2022)
- [Automatic Hardware Pragma Insertion in High-Level Synthesis: A Non-Linear Programming Approach](https://mlsanity.com/paper/2405.12304/) (2024)
- [Efficient Task Transfer for HLS DSE](https://mlsanity.com/paper/2408.13270/) (2024)
- [Accelerating Large Language Model Pretraining via LFR Pedagogy: Learn, Focus, and Review](https://mlsanity.com/paper/2409.06131/) (2024)

## Built on by (most cited first, 8 of 8)

- [iDSE: Navigating Design Space Exploration in High-Level Synthesis Using LLMs](https://mlsanity.com/paper/2505.22086/) (2025)
- [Bench4HLS: End-to-End Evaluation of LLMs in High-Level Synthesis Code Generation](https://mlsanity.com/paper/2601.19941/) (2026)
- [Agent Factories for High Level Synthesis: How Far Can General-Purpose Coding Agents Go in Hardware Optimization?](https://mlsanity.com/paper/2603.25719/) (2026)
- [Evidence-Driven LLM Agent for C-to-Synthesizable-C Conversion and Verification](https://mlsanity.com/paper/2606.28409/) (2026)
- [A3D: Agentic AI flow for autonomous Accelerator Design](https://mlsanity.com/paper/2605.15237/) (2026)
- [HLSmith: An Expert-Guided Agentic Framework for C/C++-to-HLS Translation](https://mlsanity.com/paper/2608.06791/) (2026)
- [Can Agents Design Better Chips with a Higher Level Abstraction?](https://mlsanity.com/paper/2609.21157/) (2026)
- [SyntheticHLS: Building Diverse Synthetic High-Level Synthesis Datasets using LLMs](https://mlsanity.com/paper/2610.00106/) (2026)

## 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.