mlsanityData to 14 Oct 2026

Iceberg: Enhancing HLS Modeling with Synthetic Data

arXiv 2507.09948 ↗First posted 14 Jul 2025cs.LG cs.ARcited by 1

On the map: Synthetic data and self-improvement for LLMs → LLM synthetic data generation

Abstract (arXiv)

Deep learning-based prediction models for High-Level Synthesis (HLS) of hardware designs often struggle to generalize. In this paper, we study how to close the generalizability gap of these models through pretraining on synthetic data and introduce Iceberg, a synthetic data augmentation approach that expands both large language model (LLM)-generated programs and weak labels of unseen design configurations. Our weak label generation method is integrated with an in-context model architecture, enabling meta-learning from actual and proximate labels. Iceberg improves the geometric mean modeling accuracy by 86.4% when adapt to six real-world applications with few-shot examples and achieves a 2.47× and a 1.12× better offline DSE performance when adapting to two different test datasets. Our open-sourced code is here: https://github.com/UCLA-VAST/iceberg

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in LLM synthetic data generation, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 33 of its 43 references that are in the map. First posted on arXiv 14 Jul 2025.

Builds on

33 of 43 references.

References · 33

  1. 2017Attention Is All You Needcited by 60,361
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  4. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  5. 2022TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Secondcited by 674
  6. 2018Neural Processescited by 311
  7. 2023VerilogEval: Evaluating Large Language Models for Verilog Code Generationcited by 252
  8. 2022Transformer Neural Processes: Uncertainty-Aware Meta Learning Via Sequence Modelingcited by 127
  9. 2023GPT4AIGChip: Towards Next-Generation AI Accelerator Design Automation via Large Language Modelscited by 61
  10. 2024OriGen:Enhancing RTL Code Generation with Code-to-Code Augmentation and Self-Reflectioncited by 56
  11. 2024MG-Verilog: Multi-grained Dataset Towards Enhanced LLM-assisted Verilog Generationcited by 46
  12. 2024Mixture of In-Context Prompters for Tabular PFNscited by 33
21 more
  1. 2024LICO: Large Language Models for In-Context Molecular Optimizationcited by 26
  2. 2020AutoDSE: Enabling Software Programmers to Design Efficient FPGA Acceleratorscited by 25
  3. 2024HLSPilot: LLM-based High-Level Synthesiscited by 25
  4. 2023ExPT: Synthetic Pretraining for Few-Shot Experimental Designcited by 22
  5. 2021ScaleHLS: A New Scalable High-Level Synthesis Framework on Multi-Level Intermediate Representationcited by 21
  6. 2024Automated C/C++ Program Repair for High-Level Synthesis via Large Language Modelscited by 18
  7. 2024Allo: A Programming Model for Composable Accelerator Designcited by 16
  8. 2024C2HLSC: Can LLMs Bridge the Software-to-Hardware Design Gap?cited by 12
  9. 2022PowerGear: Early-Stage Power Estimation in FPGA HLS via Heterogeneous Edge-Centric GNNscited by 10
  10. 2024Cross-Modality Program Representation Learning for Electronic Design Automation with High-Level Synthesiscited by 10
  11. 2023HIDA: A Hierarchical Dataflow Compiler for High-Level Synthesiscited by 9
  12. 2024Hierarchical Mixture of Experts: Generalizable Learning for High-Level Synthesiscited by 9
  13. 2024HLSFactory: A Framework Empowering High-Level Synthesis Datasets for Machine Learning and Beyondcited by 8
  14. 2024LLM-Aided Compilation for Tensor Acceleratorscited by 8
  15. 2024Optimizing High-Level Synthesis Designs with Retrieval-Augmented Large Language Modelscited by 6
  16. 2024Hierarchical Source-to-Post-Route QoR Prediction in High-Level Synthesis with GNNscited by 5
  17. 2024A Unified Framework for Automated Code Transformation and Pragma Insertioncited by 5
  18. 2025Stream-HLS: Towards Automatic Dataflow Accelerationcited by 5
  19. 2024Efficient Task Transfer for HLS DSEcited by 4
  20. 2024LightningSimV2: Faster and Scalable Simulation for High-Level Synthesis via Graph Compilation and Optimizationcited by 3
  21. 2024Learning to Compare Hardware Designs for High-Level Synthesiscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026SyntheticHLS: Building Diverse Synthetic High-Level Synthesis Datasets using LLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026HLSFactory-Agent: Large-Scale Agentic HLS Dataset Construction from Academic and Open-Source Projects
  2. 2025SAGE-HLS: Syntax-Aware AST-Guided LLM for High-Level Synthesis Code Generation
  3. 2024New Solutions on LLM Acceleration, Optimization, and Application
  4. 2025BARE: Leveraging Base Language Models for Few-Shot Synthetic Data Generation
  5. 2025HLSTester: Efficient Testing of Behavioral Discrepancies with LLMs for High-Level Synthesis ↗
  6. 2023Enhancing ML model accuracy for Digital VLSI circuits using diffusion models: A study on synthetic data generation
  7. 2025Can Reasoning Models Reason about Hardware? An Agentic HLS Perspective
  8. 2024SynthAI: A Multi Agent Generative AI Framework for Automated Modular HLS Design Generation