{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T18:52:09Z","timestamp":1771959129296,"version":"3.50.1"},"reference-count":38,"publisher":"Association for Computing Machinery (ACM)","issue":"PLDI","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Proc. ACM Program. Lang."],"published-print":{"date-parts":[[2025,6,10]]},"abstract":"<jats:p>\n            Domain-specific languages (DSLs) for machine learning are revolutionizing the speed and efficiency of machine learning workloads as they enable users easy access to high-performance compiler optimizations and accelerators. However, to take advantage of these capabilities, a user must first translate their legacy code from the language it is currently written in, into the new DSL. The process of automatically lifting code into these DSLs has been identified by several recent works, which propose program synthesis as a solution. However, synthesis is expensive and struggles to scale without carefully designed and hard-wired heuristics. In this paper, we present an approach for lifting that combines an enumerative synthesis approach with a Large Language Model used to\n            <jats:italic toggle=\"yes\">automatically<\/jats:italic>\n            learn the domain-specific heuristics for program lifting, in the form of a probabilistic grammar. Our approach outperforms the state-of-the-art tools in this area, despite only using\n            <jats:italic toggle=\"yes\">learned<\/jats:italic>\n            heuristics.\n          <\/jats:p>","DOI":"10.1145\/3729330","type":"journal-article","created":{"date-parts":[[2025,6,13]],"date-time":"2025-06-13T16:02:27Z","timestamp":1749830547000},"page":"1984-2006","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["Guided Tensor Lifting"],"prefix":"10.1145","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-4619-3476","authenticated-orcid":false,"given":"Yixuan","family":"Li","sequence":"first","affiliation":[{"name":"University of Edinburgh, Edinburgh, United Kingdom"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2767-1130","authenticated-orcid":false,"given":"Jos\u00e9 Wesley de Souza","family":"Magalh\u00e3es","sequence":"additional","affiliation":[{"name":"University of Edinburgh, Edinburgh, United Kingdom"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5774-3970","authenticated-orcid":false,"given":"Alexander","family":"Brauckmann","sequence":"additional","affiliation":[{"name":"University of Edinburgh, Edinburgh, United Kingdom"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1619-5052","authenticated-orcid":false,"given":"Michael F. P.","family":"O'Boyle","sequence":"additional","affiliation":[{"name":"University of Edinburgh, Edinburgh, United Kingdom"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9032-7661","authenticated-orcid":false,"given":"Elizabeth","family":"Polgreen","sequence":"additional","affiliation":[{"name":"University of Edinburgh, Edinburgh, United Kingdom"}]}],"member":"320","published-online":{"date-parts":[[2025,6,13]]},"reference":[{"key":"e_1_2_2_1_1","doi-asserted-by":"publisher","unstructured":"Mart\u00edn Abadi Ashish Agarwal Paul Barham Eugene Brevdo Zhifeng Chen Craig Citro Greg S. Corrado Andy Davis Jeffrey Dean Matthieu Devin Sanjay Ghemawat Ian Goodfellow Andrew Harp Geoffrey Irving Michael Isard Yangqing Jia Rafal Jozefowicz Lukasz Kaiser Manjunath Kudlur Josh Levenberg Dan Mane Rajat Monga Sherry Moore Derek Murray Chris Olah Mike Schuster Jonathon Shlens Benoit Steiner Ilya Sutskever Kunal Talwar Paul Tucker Vincent Vanhoucke Vijay Vasudevan Fernanda Viegas Oriol Vinyals Pete Warden Martin Wattenberg Martin Wicke Yuan Yu and Xiaoqiang Zheng. 2016. TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems. https:\/\/doi.org\/10.48550\/arXiv.1603.04467 arxiv:1603.04467. 10.48550\/arXiv.1603.04467","DOI":"10.48550\/arXiv.1603.04467"},{"key":"e_1_2_2_2_1","unstructured":"Matej Balog Alexander L. Gaunt Marc Brockschmidt Sebastian Nowozin and Daniel Tarlow. 2017. DeepCoder: Learning to Write Programs. arxiv:1611.01989. arxiv:1611.01989"},{"key":"e_1_2_2_3_1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2405.15880"},{"key":"e_1_2_2_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/3360594"},{"key":"e_1_2_2_5_1","unstructured":"Sahil Bhatia Jie Qiu Sanjit A Seshia and Alvin Cheung. 2024. Can LLMs Perform Verified Lifting of Code? Technical Report https:\/\/www2.eecs.berkeley.edu\/Pubs\/TechRpts\/2024\/EECS-2024-11.pdf"},{"key":"e_1_2_2_6_1","volume-title":"Chris Leary, Dougal Maclaurin, George Necula, Adam Paszke, Jake VanderPlas, Skye Wanderman-Milne, and Qiao Zhang.","author":"Bradbury James","year":"2018","unstructured":"James Bradbury, Roy Frostig, Peter Hawkins, Matthew James Johnson, Chris Leary, Dougal Maclaurin, George Necula, Adam Paszke, Jake VanderPlas, Skye Wanderman-Milne, and Qiao Zhang. 2018. JAX: composable transformations of Python+NumPy programs. http:\/\/github.com\/google\/jax"},{"key":"e_1_2_2_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/PACT58117.2023.00012"},{"key":"e_1_2_2_8_1","doi-asserted-by":"publisher","DOI":"10.5555\/3327144.3327180"},{"key":"e_1_2_2_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/3459010"},{"key":"e_1_2_2_10_1","unstructured":"Mehdi Drissi Olivia Watkins Aditya Khant Vivaswat Ojha Pedro Sandoval Rakia Segev Eric Weiner and Robert Keller. 2018. Program Language Translation Using a Grammar-Driven Tree-to-Tree Model. arxiv:1807.01784. arxiv:1807.01784"},{"key":"e_1_2_2_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/643470.643472"},{"key":"e_1_2_2_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/3377555.3377893"},{"key":"e_1_2_2_13_1","volume-title":"MLX: Efficient and flexible machine learning on Apple silicon. https:\/\/github.com\/ml-explore","author":"Hannun Awni","year":"2023","unstructured":"Awni Hannun, Jagrit Digani, Angelos Katharopoulos, and Ronan Collobert. 2023. MLX: Efficient and flexible machine learning on Apple silicon. https:\/\/github.com\/ml-explore"},{"key":"e_1_2_2_14_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2649-2"},{"key":"e_1_2_2_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/2980983.2908117"},{"key":"e_1_2_2_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/ASE.2017.8115709"},{"key":"e_1_2_2_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3133901"},{"key":"e_1_2_2_18_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-54862-8_26"},{"key":"e_1_2_2_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3656445"},{"key":"e_1_2_2_20_1","volume-title":"MLIR: A Compiler Infrastructure for the End of Moore\u2019s Law. arxiv:2002.11054. arxiv:2002.11054","author":"Lattner Chris","year":"2020","unstructured":"Chris Lattner, Mehdi Amini, Uday Bondhugula, Albert Cohen, Andy Davis, Jacques Pienaar, River Riddle, Tatiana Shpeisman, Nicolas Vasilache, and Oleksandr Zinenko. 2020. MLIR: A Compiler Infrastructure for the End of Moore\u2019s Law. arxiv:2002.11054. arxiv:2002.11054"},{"key":"e_1_2_2_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/3192366.3192410"},{"key":"e_1_2_2_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/3296979.3192410"},{"key":"e_1_2_2_23_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-65630-9_15"},{"key":"e_1_2_2_24_1","unstructured":"2024. llamacpp. https:\/\/github.com\/leloykun\/llama2.cpp\/ Accessed: 2024-01-19"},{"key":"e_1_2_2_25_1","doi-asserted-by":"publisher","DOI":"10.1145\/3624007.3624053"},{"key":"e_1_2_2_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/3527315"},{"key":"e_1_2_2_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/3578360.3580262"},{"key":"e_1_2_2_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3520312.3534866"},{"key":"e_1_2_2_29_1","volume-title":"Proceedings of the 36th International Conference on Machine Learning, Kamalika Chaudhuri and Ruslan Salakhutdinov (Eds.) (Proceedings of Machine Learning Research","volume":"4870","author":"Nye Maxwell","year":"2019","unstructured":"Maxwell Nye, Luke Hewitt, Joshua Tenenbaum, and Armando Solar-Lezama. 2019. Learning to Infer Program Sketches. In Proceedings of the 36th International Conference on Machine Learning, Kamalika Chaudhuri and Ruslan Salakhutdinov (Eds.) (Proceedings of Machine Learning Research, Vol. 97). PMLR, 4861\u20134870. https:\/\/proceedings.mlr.press\/v97\/nye19a.html"},{"key":"e_1_2_2_30_1","doi-asserted-by":"publisher","DOI":"10.1006\/jpdc.2001.1815"},{"key":"e_1_2_2_31_1","volume-title":"BUSTLE: Bottom-Up Program Synthesis Through Learning-Guided Exploration. arxiv:2007.14381. arxiv:2007.14381","author":"Odena Augustus","year":"2021","unstructured":"Augustus Odena, Kensen Shi, David Bieber, Rishabh Singh, Charles Sutton, and Hanjun Dai. 2021. BUSTLE: Bottom-Up Program Synthesis Through Learning-Guided Exploration. arxiv:2007.14381. arxiv:2007.14381"},{"key":"e_1_2_2_32_1","volume-title":"The Twelfth International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=y0GJXRungR","author":"Olausson Theo X.","year":"2024","unstructured":"Theo X. Olausson, Jeevana Priya Inala, Chenglong Wang, Jianfeng Gao, and Armando Solar-Lezama. 2024. Is Self-Repair a Silver Bullet for Code Generation? In The Twelfth International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=y0GJXRungR"},{"key":"e_1_2_2_33_1","unstructured":"OpenAI. 2024. GPT-4 Technical Report. arxiv:2303.08774. arxiv:2303.08774"},{"key":"e_1_2_2_34_1","first-page":"8026","article-title":"PyTorch: An Imperative Style, High-Performance Deep Learning Library","author":"Paszke Adam","year":"2019","unstructured":"Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas Kopf, Edward Yang, Zachary DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu Fang, Junjie Bai, and Soumith Chintala. 2019. PyTorch: An Imperative Style, High-Performance Deep Learning Library. In Advances in Neural Information Processing Systems. 32, Curran Associates, Inc., 8026\u20138037. isbn:978-1-71380-793-3 https:\/\/dl.acm.org\/doi\/10.5555\/3454287.3455008","journal-title":"Advances in Neural Information Processing Systems. 32, Curran Associates, Inc."},{"key":"e_1_2_2_35_1","doi-asserted-by":"publisher","DOI":"10.4230\/LIPIcs.ECOOP.2024.32"},{"key":"e_1_2_2_36_1","doi-asserted-by":"publisher","DOI":"10.1145\/2491956.2462176"},{"key":"e_1_2_2_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3517034"},{"key":"e_1_2_2_38_1","unstructured":"Kensen Shi Hanjun Dai Kevin Ellis and Charles Sutton. 2022. CrossBeam: Learning to Search in Bottom-Up Program Synthesis. arxiv:2203.10452. arxiv:2203.10452"}],"container-title":["Proceedings of the ACM on Programming Languages"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3729330","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,16]],"date-time":"2025-07-16T06:04:31Z","timestamp":1752645871000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3729330"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,10]]},"references-count":38,"journal-issue":{"issue":"PLDI","published-print":{"date-parts":[[2025,6,10]]}},"alternative-id":["10.1145\/3729330"],"URL":"https:\/\/doi.org\/10.1145\/3729330","relation":{},"ISSN":["2475-1421"],"issn-type":[{"value":"2475-1421","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,10]]},"assertion":[{"value":"2024-11-15","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-06","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-06-13","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}