{"ID":5938064,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-07T23:38:28.215472405Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.04058","arxiv_id":"2607.04058","title":"Kaizen: Metamorphic Fuzzing and Differential Testing for LLM-Translated HPC Applications","abstract":"Large language models (LLMs) are increasingly used to port scientific codes across heterogeneous high-performance computing (HPC) programming models, such as translating CUDA to OpenMP, OpenACC, Kokkos or SYCL. However, current evaluations use compilation success, token-level similarity, or developer-written tests from static benchmarks, which cannot reliably ensure behavioral correctness. We present Kaizen, a metamorphic fuzzing and differential testing framework for evaluating the correctness of LLM-translated HPC code. Kaizen uses metamorphic fuzzing via source-code mutation to generate semantically equivalent programs, grammar-based input fuzzing to explore behavioral diversity, and differential testing to expose semantic divergences between original and translated applications that compile and pass developer-written tests yet produce incorrect scientific results. We evaluate Kaizen on CUDA-to-OpenMP translation of 16 scientific applications from seven domains using three fine-tuned LLMs at kernel-level and full-program granularity. Our evaluation reveals that (1) compilation success is a poor proxy for correctness; (2) LLM-translated programs exhibit systematic compile-time error patterns, with nine categories for kernel-level translation and 27 for full-program translation; (3) semantic errors that survive compilation are often input-dependent and require differential testing to expose; and (4) full-program translation is substantially harder than kernel-level translation. These findings highlight the need for correctness-oriented evaluation of LLM-assisted HPC code translations.","short_abstract":"Large language models (LLMs) are increasingly used to port scientific codes across heterogeneous high-performance computing (HPC) programming models, such as translating CUDA to OpenMP, OpenACC, Kokkos or SYCL. However, current evaluations use compilation success, token-level similarity, or developer-written tests from...","url_abs":"https://arxiv.org/abs/2607.04058","url_pdf":"https://arxiv.org/pdf/2607.04058v1","authors":"[\"Oscar Ludwig\",\"Ninad Anklesaria\",\"Zheming Jin\",\"Swaroop Pophale\",\"Kausar Moshood\",\"Christian J. DeVore\",\"Brandon Gill\",\"Cassius Villareal\",\"Keita Teranishi\",\"Manish Motwani\"]","published":"2026-07-04T23:52:39Z","proceeding":"cs.SE","tasks":"[\"cs.SE\",\"cs.PL\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}