mlsanityData to 7 Oct 2026

ParaCodex: A Profiling-Guided Autonomous Coding Agent for Reliable Parallel Code Generation and Translation

arXiv 2601.04327 ↗First posted 7 Jan 2026cs.DC cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Parallel programming is central to HPC and AI, but producing code that is correct and fast remains challenging, especially for OpenMP GPU offload, where data movement and tuning dominate. Autonomous coding agents can compile, test, and profile on target hardware, but outputs are brittle without domain scaffolding. We present ParaCodex, an HPC-engineer workflow that turns a Codex-based agent into an autonomous OpenMP GPU offload system using staged hotspot analysis, explicit data planning, correctness gating, and profiling-guided refinement. We evaluate translation from serial CPU kernels to OpenMP GPU offload kernels on HeCBench, Rodinia, and NAS. After excluding five kernels, ParaCodex succeeded on all 31 valid kernels. The generated kernels improved GPU time over reference OpenMP implementations in 25/31 cases, achieving geometric-mean speedups of 3x on HeCBench and 5x on Rodinia, and outperforming a zero-shot Codex baseline on all suites. We also evaluate CUDA to OpenMP offload translation on ParEval, where ParaCodex maintains high compilation and validation rates in code-only and end-to-end settings.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 33 of its 50 references that are in the map. First posted on arXiv 7 Jan 2026.

Builds on

33 of 50 references.

References · 33

  1. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  2. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  3. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  4. 2023CodeGeeX: A Pre-Trained Model for Code Generation with Multilingual Benchmarking on HumanEval-Xcited by 294
  5. 2018Learning to Optimize Tensor Programscited by 138
  6. 2023Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generationcited by 128
  7. 2024Meta Large Language Model Compiler: Foundation Models of Compiler Optimizationcited by 39
  8. 2024Can Large Language Models Write Parallel Code?cited by 36
  9. 2023Comparing Llama-2 and GPT-3 LLMs for HPC kernels generationcited by 30
  10. 2022Hidet: Task-Mapping Programming Paradigm for Deep Learning Tensor Programscited by 28
  11. 2023Evaluation of OpenAI Codex for HPC Parallel Programming Models Kernel Generationcited by 22
  12. 2024The Landscape and Challenges of HPC Research and LLMscited by 22
21 more
  1. 2024PerfCodeGen: Improving Performance of LLM Generated Code with Execution Feedbackcited by 22
  2. 2025Agentic Refactoring: An Empirical Study of AI Coding Agentscited by 17
  3. 2023Learning to Parallelize with OpenMP by Augmented Heterogeneous AST Representationcited by 14
  4. 2018The Three Pillars of Machine Programmingcited by 12
  5. 2023MonoCoder: Domain-Specific Code Language Model for HPC Codes and Taskscited by 12
  6. 2022Learning to Parallelize in a Shared-Memory Environment with Transformerscited by 10
  7. 2023Advising OpenMP Parallelization via a Graph-Based Approach with Transformerscited by 10
  8. 2024LASSI: An LLM-based Automated Self-Correcting Pipeline for Translating Parallel Scientific Codescited by 10
  9. 2019Reinforcement Learning and Adaptive Sampling for Optimized DNN Compilationcited by 9
  10. 2023Quantifying OpenMP: Statistical Insights into Usage and Adoptioncited by 9
  11. 2020ComPar: Optimized Multi-Compiler for Automatic OpenMP S2S Parallelizationcited by 8
  12. 2024LOOPer: A Learned Automatic Code Optimizer For Polyhedral Compilerscited by 7
  13. 2025Do Large Language Models Understand Performance Optimization?cited by 6
  14. 2023MPI-rical: Data-Driven MPI Distributed Parallelism Assistance with Transformerscited by 5
  15. 2024MPIrigen: MPI Code Generation through Domain-Specific Language Modelscited by 5
  16. 2024OMPar: Automatic Parallelization with AI-Driven Source-to-Source Compilationcited by 5
  17. 2025Can Large Language Models Predict Parallel Code Performance?cited by 4
  18. 2025UniPar: A Unified LLM-Based Framework for Parallel and Accelerated Code Translation in HPCcited by 4
  19. 2015Polly's Polyhedral Scheduling in the Presence of Reductionscited by 3
  20. 2025LLM Priors for ERM over Programscited by 3
  21. 2025Beyond Code Pairs: Dialogue-Based Data Generation for LLM Code Translationcited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Agentic Kernel Optimization: Generating State-of-the-Art GPU Kernels Without Hand-Written CUDA
  2. 2026A Two-Stage GPU Kernel Tuner Combining Semantic Refactoring and Search-Based Optimization
  3. 2026Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization
  4. 2026Are LLM-Generated GPU Kernels Production-Ready? A Trace-Driven Benchmark and Optimization Agent
  5. 2026AgentKernelArena: Generalization-Aware Benchmarking of GPU Kernel Optimization Agents
  6. 2026FastKernels: Benchmarking GPU Kernel Generation in Production
  7. 2026ParBench: A Benchmark for Reliable Evaluation of LLM Parallel Code Translation
  8. 2026Ave: Guiding Agentic GPU Optimization Using Data-Flow Invariants