T-LLM Compiler: Trusted LLM-based Code Optimization and Verification Framework

2026-08-14

AI Summary

This study addresses the low correctness and lack of independent verification in large language model-based code optimization by proposing a trustworthy optimization paradigm that synergizes LLMs with traditional compilers and formal verification. By constructing an optimization framework integrating iterative verification with error-correction feedback, this approach effectively ensures both transformation correctness and performance efficiency. Experimental evaluations on the PolyBench/C benchmark demonstrate that the proposed method achieves an optimization accuracy of 83.3% and an average speedup of 26.7%. These results significantly enhance the reliability and practicality of AI-driven code optimization, offering novel insights for developing trustworthy intelligent compilation systems.

Abstract

Recent advances in Large Language Models (LLMs) have opened opportunities to apply high-level code transformations to the field of code optimization, and it has since emerged as one of the most fundamental tasks for LLMs to perform; however, at present, LLMs struggle to apply wide-ranging code optimization tasks due to both the complexity of the code and the inability to independently verify the correctness of the transformations. In this paper, we present the Trusted LLM (T-LLM) Compiler, which proposes an advancement in compiler technology through a collaborative effort involving high-level LLM code transformations, traditional compilers, and verification tools. Experimental results reveal that it can significantly improve code correctness when tested on a set of PolyBench/C benchmarks. Our approach facilitates iterative code optimization efforts with verification strategies that enable corrective actions. Through this approach, T-LLM Compiler achieves code optimization accuracy of up to 83.3% and a speedup of up to 16.1\% on the PolyBench/C benchmarks, with the transformed code reaching an average of 26.7% speedup wrt standard baselines. Additionally, we release the project's source code to the open-source community.

Problem

• Code Optimization
• Large Language Models
• Correctness Verification
• Compiler

Innovation

• Trusted LLM Compiler
• Code Optimization
• Verification Framework
• Iterative Correction
• Collaborative Compilation