mlsanityData to 7 Oct 2026

MLGO: a Machine Learning Guided Compiler Optimizations Framework

arXiv 2101.04808 ↗First posted 13 Jan 2021cs.PL cs.LGcited by 40

Not on the map. Why?

Abstract (arXiv)

Leveraging machine-learning (ML) techniques for compiler optimizations has been widely studied and explored in academia. However, the adoption of ML in general-purpose, industry strength compilers has yet to happen. We propose MLGO, a framework for integrating ML techniques systematically in an industrial compiler -- LLVM. As a case study, we present the details and results of replacing the heuristics-based inlining-for-size optimization in LLVM with machine learned models. To the best of our knowledge, this work is the first full integration of ML in a complex compiler pass in a real-world setting. It is available in the main LLVM repository. We use two different ML algorithms: Policy Gradient and Evolution Strategies, to train the inlining-for-size model, and achieve up to 7% size reduction, when compared to state of the art LLVM -Oz. The same model, trained on one corpus, generalizes well to a diversity of real-world targets, as well as to the same set of targets after months of active development. This property of the trained models is beneficial to deploy ML techniques in real-world settings.

Key facts. Cited by 40 papers in the map (data through 7 Oct 2026). Builds on 8 of its 33 references that are in the map. First posted on arXiv 13 Jan 2021.

Builds on

8 of 33 references.

References · 8

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2017Inductive Representation Learning on Large Graphscited by 6,284
  3. 2013Playing Atari with Deep Reinforcement Learningcited by 3,211
  4. 2017Evolution Strategies as a Scalable Alternative to Reinforcement Learningcited by 555
  5. 2018code2vec: Learning Distributed Representations of Codecited by 249
  6. 2018Learning to Optimize Tensor Programscited by 138
  7. 2018Structured Evolution with Compact Architectures for Scalable Policy Optimizationcited by 48
  8. 2019NeuroVectorizer: End-to-End Vectorization with Deep Reinforcement Learningcited by 37

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 6; 2023: 7; 2024: 8; 2025: 6; 2026: 9.

40 papers.

Sort:

By area:

  1. 2022Reincarnating Reinforcement Learning: Reusing Prior Computation to Accelerate Progresscited by 63
  2. 2023Large Language Models for Compiler Optimizationcited by 52
  3. 2024Efficient Expert Pruning for Sparse Mixture-of-Experts Language Models: Enhancing Performance and Reducing Inference Costscited by 51
  4. 2021CompilerGym: Robust, Performant Compiler Optimization Environments for AI Researchcited by 41
  5. 2024Meta Large Language Model Compiler: Foundation Models of Compiler Optimizationcited by 39
  6. 2025STARK: Strategic Team of Agents for Refining Kernelscited by 23
  7. 2022Compiler Optimization for Quantum Computing Using Reinforcement Learningcited by 11
  8. 2023Learning Compiler Pass Orders using Coreset and Normalized Value Predictioncited by 11
  9. 2025Regularized Langevin Dynamics for Combinatorial Optimizationcited by 10
  10. 2022MLGOPerf: An ML Guided Inliner to Optimize Performancecited by 9
  11. 2022GRANITE: A Graph Neural Network Model for Basic Block Throughput Estimationcited by 9
  12. 2024Compiler generated feedback for Large Language Modelscited by 9
28 more
  1. 2024SORREL: Suboptimal-Demonstration-Guided Reinforcement Learning for Learning to Branchcited by 9
  2. 2024CompilerDream: Learning a Compiler World Model for General Code Optimizationcited by 6
  3. 2021A Reinforcement Learning Environment for Polyhedral Optimizationscited by 5
  4. 2022Learning to Combine Instructions in LLVM Compilercited by 4
  5. 2022RL4ReAl: Reinforcement Learning for Register Allocationcited by 4
  6. 2021Generating GPU Compiler Heuristics using Reinforcement Learningcited by 3
  7. 2025REASONING COMPILER: LLM-Guided Optimizations for Efficient Model Servingcited by 3
  8. 2026Agentic Harness for Real-World Compilerscited by 3
  9. 2021Profile Guided Optimization without Profiles: A Machine Learning Approachcited by 2
  10. 2023Power Constrained Autotuning using Graph Neural Networkscited by 2
  11. 2024Offline Imitation Learning from Multiple Baselines with Applications to Compiler Optimizationcited by 2
  12. 2023Target-independent XLA optimization using Reinforcement Learningcited by 1
  13. 2023LoopTune: Optimizing Tensor Computations with Reinforcement Learningcited by 1
  14. 2024Towards LLM-based optimization compilers. Can LLMs learn how to apply a single peephole optimization? Reasoning is all LLMs need!cited by 1
  15. 2025VeriLocc: End-to-End Cross-Architecture Register Allocation via LLMcited by 1
  16. 2026Protean Compiler: An Agile Framework to Drive Fine-grain Phase Orderingcited by 1
  17. 2026AI-PROPELLER: Warehouse-Scale Interprocedural Code Layout Optimization with AlphaEvolvecited by 1
  18. 2026AutoPass: Evidence-Guided LLM Agents for Compiler Performance Tuningcited by 1
  19. 2023ML-driven Hardware Cost Model for MLIRcited by 0
  20. 2023The Next 700 ML-Enabled Compiler Optimizationscited by 0
  21. 2024A Research Agenda for Usability and Generalisation in Reinforcement Learningcited by 0
  22. 2025QiMeng-NeuComBack: Self-Evolving Translation from IR to Assembly Codecited by 0
  23. 2025A Deep Learning Model for Predicting Transformation Legalitycited by 0
  24. 2026ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evolutioncited by 0
  25. 2026Can Large Language Models Recover Semantic Optimization Opportunities That Compilers Miss?cited by 0
  26. 2026CompileRover: Revolutionizing Virtual Machine Compiler Optimization with a Tri-Role LLM-Driven Frameworkcited by 0
  27. 2026MQSS-Selector: RL-Guided Pass Selection for an MLIR Compilation Pipelinecited by 0
  28. 2026Self-Evolving Coding Agents: From Digital Programs to Physical-World Intelligencecited by 0