Beyond Pass-by-Pass Optimization: Intent-Driven IR Optimization with Large Language Models

2026-02-19

AI Summary

Traditional compilers perform optimization in separate passes without global intent, often yielding suboptimal results. While recent large language model (LLM)-based approaches generate intermediate representation (IR) code end-to-end, they fail to explicitly model optimization intent, compromising both correctness and performance. This work proposes IntOpt, the first IR optimization framework that explicitly decouples high-level optimization intent from low-level transformations. IntOpt structures the optimization process into three stages—intent formulation, refinement, and implementation—integrating program analysis, intent reasoning, and LLM-driven code generation to achieve globally coordinated and structured optimizations. Evaluated on 200 programs, IntOpt achieves a 90.5% verification correctness rate and an average speedup of 2.66×, outperforming state-of-the-art LLM-based optimizers and surpassing -O3 on 37 benchmarks, with a peak speedup of 272.60×.

Technology Category

• Machine Learning: Optimization
• Search and Optimization: Local Search
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Modern compilers optimize programs through a sequence of modular passes over intermediate representations (IR). While this pass-by-pass paradigm offers engineering benefits, it suffers from a pass coordination problem: locally beneficial transformations may block more profitable optimizations in later stages. This limitation stems from the lack of an explicit notion of optimization intent, defined as a holistic strategy for coordinating multiple transformations toward a global performance objective. Recent LLM-based approaches formulate IR optimization as an end-to-end generation task, thereby avoiding the traditional pass-by-pass structure. However, optimization intent remains implicit in these methods, forcing models to jointly infer optimization strategy and generate low-level transformations, which limits both correctness and performance. We propose IntOpt, the first intent-driven IR optimizer that explicitly separates high-level optimization intent from low-level analysis and transformation. IntOpt organizes IR optimization into three stages: intent formulation, intent refinement, and intent realization, enabling globally coordinated transformations. Experiments show that IntOpt achieves 90.5% verified correctness and 2.660x average speedup on 200-program test set, outperforming state-of-the-art LLM-based optimizers in both correctness and performance, and surpassing modern compiler with the -O3 option on 37 benchmarks with speedups of up to 272.60x.

Problem

• IR optimization
• optimization intent
• pass coordination
• compiler optimization
• large language models

Innovation

• intent-driven optimization
• intermediate representation (IR)
• large language models
• compiler optimization
• global coordination