{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T09:02:40Z","timestamp":1783069360156,"version":"3.54.6"},"reference-count":43,"publisher":"Association for Computing Machinery (ACM)","issue":"6","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Des. Autom. Electron. Syst."],"published-print":{"date-parts":[[2025,11,30]]},"abstract":"<jats:p>Deep learning tasks, especially those involving complex convolution neural networks (CNNs), are computationally intensive and pose significant challenges when implemented on hardware. Accelerating these tasks is critical for improving performance. High-level Synthesis (HLS) has the potential to automate the efficient hardware accelerator designs directly from high-level C\/C++ specification of trained machine learning (ML) models. Traditional HLS tools cannot synthesize certain high-level constructs, which require manual intervention. Many source code optimizations and the selection of pragmas for HLS optimizations are crucial for generating efficient hardware accelerators with HLS. However, both of these tasks are mostly manual efforts. Recently, Large Language Models (LLMs) have shown remarkable capabilities in various generative tasks. In this work, we explore the application of LLMs to remove these manual efforts in adapting HLS for ML accelerator designs. Our framework called LLM-assisted HLS, i.e., LHS, uses LLMs to automate the resolution of synthesis issues, ensuring compatibility with HLS tools. Furthermore, our framework automates the source code modification and optimization selection through pragma insertion steps, which are crucial for optimizing the synthesized design. Our experimental results with LHS demonstrate a significant improvement in latency for deep learning tasks with underlying complex CNN models without much area overhead. Our LHS allows us to achieve up to 2690 \u00d7 latency improvement. Promisingly, LHS performs better than the state-of-the-art ML accelerator design tool hls4ml in 4 out of 6 cases in the context of latency improvement at the expense of area overhead (i.e., performance to hardware gain). This work highlights the potential of LLMs to assist and accelerate the HLS process, thereby creating more efficient hardware implementation for deep learning models.<\/jats:p>","DOI":"10.1145\/3734523","type":"journal-article","created":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T07:06:36Z","timestamp":1746687996000},"page":"1-27","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["LHS: LLM Assisted Efficient High-level Synthesis of Deep Learning Tasks"],"prefix":"10.1145","volume":"30","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-5116-0971","authenticated-orcid":false,"given":"E Bhawani Eswar","family":"Reddy","sequence":"first","affiliation":[{"name":"Computer Science and Engineering, Indian Institute of Technology Guwahati","place":["Guwahati, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-4786-8627","authenticated-orcid":false,"given":"Sutirtha","family":"Bhattacharyya","sequence":"additional","affiliation":[{"name":"Indian Association for the Cultivation of Science","place":["Kolkata, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-9705-4564","authenticated-orcid":false,"given":"Ankur","family":"Sarmah","sequence":"additional","affiliation":[{"name":"Tezpur University","place":["Napaam, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-8566-3519","authenticated-orcid":false,"given":"Fedrick","family":"Nongpoh","sequence":"additional","affiliation":[{"name":"Nagaland University","place":["Lumami, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-3395-8663","authenticated-orcid":false,"given":"Karthik","family":"Maddala","sequence":"additional","affiliation":[{"name":"Computer Science and Engineering, Indian Institute of Technology Guwahati","place":["Guwahati, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3835-4184","authenticated-orcid":false,"given":"Chandan","family":"Karfa","sequence":"additional","affiliation":[{"name":"Computer Science and Engineering, Indian Institute of Technology Guwahati","place":["Guwahati, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2025,10,21]]},"reference":[{"key":"e_1_3_2_2_2","first-page":"265","volume-title":"Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI\u201916)","author":"Abadi Mart\u00edn","year":"2016","unstructured":"Mart\u00edn Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker, Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng2016. \\(\\lbrace\\) TensorFlow \\(\\rbrace\\) : A system for \\(\\lbrace\\) Large-Scale \\(\\rbrace\\) machine learning. In Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI\u201916). 265\u2013283."},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1145\/197405.197406"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1145\/3670474.3685948"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","unstructured":"K. Chang K. Wang N. Yang Y. Wang D. Jin W. Zhu Z. Chen C. Li H. Yan Y. Zhou Z. Zhao Y. Cheng Y. Pan Y. Liu M. Wang S. Liang Y. Han H. Li and X. Li. 2024. Data is All you Need: Finetuning LLMs for Chip Design via an Automated Design-data Augmentation Framework. In Proceedings of the 61st ACM\/IEEE Design Automation Conference (DAC\u201924). Association for Computing Machinery New York NY USA Article 60 1\u20136. 10.1145\/3649329.3657356","DOI":"10.1145\/3649329.3657356"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","unstructured":"L. Chen Y. Chen Z. Chu et\u00a0al. 2024. Large circuit models: Opportunities and challenges. Sci. China Inf. Sci. 67 200402 (2024). 10.1007\/s11432-024-4155-7","DOI":"10.1007\/s11432-024-4155-7"},{"key":"e_1_3_2_7_2","unstructured":"Fran\u00e7ois Chollet et\u00a0al. 2015. Keras. Retrieved April 30 2025 from https:\/\/github.com\/fchollet\/keras"},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","unstructured":"L. Collini S. Garg and R. Karri. 2024. C2HLSC: Can LLMs Bridge the Software-to-Hardware Design Gap?IEEE LLM Aided Design Workshop (LAD) San Jose CA USA 2024. 1\u201312. DOI:10.1109\/LAD62341.2024.10691856","DOI":"10.1109\/LAD62341.2024.10691856"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","unstructured":"Rory Conlin Keith Erickson Joseph Abbate and Egemen Kolemen. 2021. Keras2c: A library for converting Keras neural networks to real-time compatible C. Engineering Applications of Artificial Intelligence 100 2021 (2021) 104182. 10.1016\/j.engappai.2021.104182","DOI":"10.1016\/j.engappai.2021.104182"},{"key":"e_1_3_2_10_2","unstructured":"Shukai Duan Nikos Kanakaris Xiongye Xiao Heng Ping Chenyu Zhou Nesreen K. Ahmed Guixiang Ma Mihai Capota Theodore L. Willke Shahin Nazarian and Paul Bogdan. 2023. Leveraging Reinforcement Learning and Large Language Models for Code Optimization. [cs.LG] arXiv:2312.05657. Retrieved from https:\/\/arxiv.org\/abs\/2312.05657"},{"key":"e_1_3_2_11_2","unstructured":"Farah Fahim Benjamin Hawks Christian Herwig James Hirschauer Sergo Jindariani Nga Viet Tran Luca P. Carloni Giuseppe Di Guglielmo Philip Harris Jeffrey Krupa Dylan Rankin Manuel Blanco Valentin Josiah Hester Yingyi Luo John Mamish Seda Orgrenci-Memik Thea Aarestaad Hamza Javed Vladimir Loncar Maurizio Pierini Adrian Alan Pol Sioni Summers Javier Mauricio Duarte Scott Hauck Shih-Chieh Hsu Jennifer Ngadiuba Mia Liu Duc Hoang Edward Kreinar and Zhenbin Wu. 2021. hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devices. [cs.LG] arXiv:2103.05579. Retrieved from https:\/\/arxiv.org\/abs\/2103.05579"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICSE48619.2023.00128"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1145\/3660773"},{"key":"e_1_3_2_14_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-57077-4_10"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","unstructured":"Rahul Kande Hammond Pearce Benjamin Tan Brendan Dolan-Gavitt Shailja Thakur Ramesh Karri and Jeyavijayan Rajendran. 2024. (Security) assertions by large language models. In IEEE Transactions on Information Forensics and Security 19 (2024) 4374\u20134389. DOI:10.1109\/TIFS.2024.3372809","DOI":"10.1109\/TIFS.2024.3372809"},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4842-2766-4_7"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2834439"},{"key":"e_1_3_2_18_2","doi-asserted-by":"crossref","unstructured":"Wen Lai Viktor Hangya and Alexander Fraser. 2024. Style-Specific Neurons for Steering LLMs in Text Style Transfer. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. Miami Florida USA. Association for Computational Linguistics 13427\u201313443.","DOI":"10.18653\/v1\/2024.emnlp-main.745"},{"key":"e_1_3_2_19_2","unstructured":"Mingjie Liu Teodor-Dumitru Ene Robert Kirby Chris Cheng Nathaniel Pinckney Rongjian Liang Jonah Alben Himyanshu Anand Sanmitra Banerjee Ismet Bayraktaroglu Bonita Bhaskaran Bryan Catanzaro Arjun Chaudhuri Sharon Clay Bill Dally Laura Dang Parikshit Deshpande Siddhanth Dhodhi Sameer Halepete Eric Hill Jiashang Hu Sumit Jain Ankit Jindal Brucek Khailany George Kokai Kishor Kunal Xiaowei Li Charley Lind Hao Liu Stuart Oberman Sujeet Omar Ghasem Pasandi Sreedhar Pratty Jonathan Raiman Ambar Sarkar Zhengjiang Shao Hanfei Sun Pratik P Suthar Varun Tej Walker Turner Kaizhe Xu and Haoxing Ren. 2024. ChipNeMo: Domain-Adapted LLMs for Chip Design. [cs.CL] arXiv:2311.00176. Retrieved from https:\/\/arxiv.org\/abs\/2311.00176"},{"key":"e_1_3_2_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD57390.2023.10323812"},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1109\/ITCIndia62949.2024.10651860"},{"key":"e_1_3_2_22_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISVLSI61997.2024.00130"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/DSD57027.2022.00019"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1145\/3670474.3685964"},{"key":"e_1_3_2_25_2","unstructured":"OpenAI. 2023. ChatGPT: A Language Model by OpenAI. Retrieved April 30 2025 from https:\/\/chat.openai.com\/"},{"key":"e_1_3_2_26_2","unstructured":"S. Qiu B. Tan and H. Pearce. 2024. Explaining EDA Synthesis Errors with LLMs. arxiv:2404.07235. Retrieved from https:\/\/arxiv.org\/abs\/2404.07235"},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.23919\/DATE56975.2023.10137086"},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3657353"},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1145\/3670474.3685966"},{"key":"e_1_3_2_30_2","first-page":"435","volume-title":"Proceedings of the 2024 29th ASP-DAC","author":"Wan L. J.","year":"2024","unstructured":"L. J. Wan, Y. Huang, Y. Li, H. Ye, J. Wang, X. Zhang, and D. Chen. 2024. Software\/hardware co-design for llm and its application for design verification. In Proceedings of the 2024 29th ASP-DAC. IEEE, 435\u2013441."},{"key":"e_1_3_2_31_2","volume-title":"Electronic Design Automation: Synthesis, Verification, and Test","author":"Wang Laung-Terng","year":"2009","unstructured":"Laung-Terng Wang, Yao-Wen Chang, and Kwang-Ting Tim Cheng. 2009. Electronic Design Automation: Synthesis, Verification, and Test. Morgan Kaufmann."},{"key":"e_1_3_2_32_2","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3658493"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISVLSI61997.2024.00076"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2024.3383347"},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","unstructured":"Z. Xi W. Chen X. Guo W. He Y. Ding B. Hong M. Zhang J. Wang S. Jin E. Zhou R. Zheng X. Fan X. Wang L. Xiong Y. Zhou W. Wang C. Jiang Y. Zou X. Liu Z. Yin S. Dou R. Weng W. Cheng Q. Zhang W. Qin Y. Zheng X. Qiu X. Huang and T. Gui. 2025. The Rise and Potential of Large Language Model Based Agents: A Survey. Sci. China Inf. Sci. 68 121101 (2025). 10.1007\/s11432-024-4222-0","DOI":"10.1007\/s11432-024-4222-0"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2402.09099"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-48981-x"},{"key":"e_1_3_2_38_2","unstructured":"Xilinx. 2012. Vivado HLS. Retrieved June 12 2024 from http:\/\/xilinx.com\/support\/download.html"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","unstructured":"Chenwei Xiong Cheng Liu Huawei Li and Xiaowei Li. 2025. HLSPilot: LLM-based High-Level Synthesis. In Proceedings of the 43rd IEEE\/ACM International Conference on Computer-Aided Design (ICCAD\u201924). Association for Computing Machinery New York NY USA Article 226 1\u20139. 10.1145\/3676536.3676781","DOI":"10.1145\/3676536.3676781"},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","unstructured":"Kangwei Xu Grace Li Zhang Xunzhao Yin Cheng Zhuo Ulf Schlichtmann and Bing Li. 2024. Automated C\/C++ Program Repair for High-Level Synthesis via Large Language Models. In Proceedings of the 2024 ACM\/IEEE International Symposium on Machine Learning for CAD (MLCAD\u201924). Association for Computing Machinery New York NY USA Article 15 1\u20139. 10.1145\/3670474.3685953","DOI":"10.1145\/3670474.3685953"},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1109\/ASE56229.2023.00096"},{"key":"e_1_3_2_42_2","doi-asserted-by":"publisher","DOI":"10.1145\/3676536.3676816"},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3082304"},{"key":"e_1_3_2_44_2","unstructured":"R. Zhong X. Du S. Kai Z. Tang S. Xu H.-L. Zhen J. Hao Q. Xu M. Yuan and J. Yan. 2023. LLM4EDA: Emerging Progress in Large Language Models for Electronic Design Automation. [cs.AR] arxiv:2401.12224. Retrieved from https:\/\/arxiv.org\/abs\/2401.12224"}],"container-title":["ACM Transactions on Design Automation of Electronic Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3734523","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T12:21:08Z","timestamp":1761135668000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3734523"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,21]]},"references-count":43,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,11,30]]}},"alternative-id":["10.1145\/3734523"],"URL":"https:\/\/doi.org\/10.1145\/3734523","relation":{},"ISSN":["1084-4309","1557-7309"],"issn-type":[{"value":"1084-4309","type":"print"},{"value":"1557-7309","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,21]]},"assertion":[{"value":"2024-09-30","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-04-21","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-10-21","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}