Compiler generated feedback for Large Language Models
원문 초록
We introduce a novel paradigm in compiler optimization powered by Large Language Models with compiler feedback to optimize the code size of LLVM assembly. The model takes unoptimized LLVM IR as input and produces optimized IR, the best optimization passes, and instruction counts of both unoptimized and optimized IRs. Then we compile the input with generated optimization passes and evaluate if the predicted instruction count is correct, generated IR is compilable, and corresponds to compiled code. We provide this feedback back to LLM and give it another chance to optimize code. This approach adds an extra 0.53% improvement over -Oz to the original model. Even though, adding more information with feedback seems intuitive, simple sampling techniques achieve much higher performance given 10 or more samples.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Bridging the PLC Binary Analysis Gap: A Cross-Compiler Dataset and Neural Framework for Industrial Control Systems
Optimization of Armv9 architecture general large language model inference performance based on Llama.cpp
WhiteFox: White-Box Compiler Fuzzing Empowered by Large Language Models
Compiler Optimization for Quantum Computing Using Reinforcement Learning
BaCO: A Fast and Portable Bayesian Compiler Optimization Framework
SimCLF: A Simple Contrastive Learning Framework for Function-level Binary Embeddings
토론
이 논문과 연결된 포럼 글이 아직 없습니다.