REVIEW 4 cited by
Performance-Aligned LLMs for Generating Fast Code
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Optimizing scientific software is a difficult task because codebases are often large and complex, and performance can depend upon several factors including the algorithm, its implementation, and hardware among others. Causes of poor performance can originate from disparate sources and be difficult to diagnose. Recent years have seen a multitude of work that use large language models (LLMs) to assist in software development tasks. However, these tools are trained to model the distribution of code as text, and are not specifically designed to understand performance aspects of code. In this work, we introduce a reinforcement learning based methodology to align the outputs of code LLMs with performance. This allows us to build upon the current code modeling capabilities of LLMs and extend them to generate better performing code. We demonstrate that our fine-tuned model improves the expected speedup of generated code over base models for a set of benchmark tasks from 0.9 to 1.6 for serial code and 1.9 to 4.5 for OpenMP code.
Forward citations
Cited by 4 Pith papers
-
Effect of Abstractions and Prompting Strategies on LLM-Guided High-Performance Optimizations
Providing LLMs with specific optimization hints and letting them generate plain C code outperforms using Halide, Exo, and Noarr abstractions for PolyBench optimization in the tested non-iterative setting.
-
HPC-Coder-V2: Studying Code LLMs Across Low-Resource Parallel Languages
Fine-tuning DeepSeek-Coder on a new 122k synthetic HPC instruction dataset yields open-source models that reach 34.1 pass@1 on ParEval parallel code generation, besting other open baselines but trailing GPT-4.
-
Can Large Language Models Predict Parallel Code Performance?
The best LLMs (o1, o3-mini-high) classify GPU kernels as compute- or bandwidth-bound from source with 64% accuracy, versus 100% when given profiled arithmetic intensity.
-
Language Models for Code Optimization: Survey, Challenges and Future Directions
A systematic review of 53 papers on using large language models for code optimization, with a taxonomy, five challenges, and eight future research directions.
Discussion (0). Continue with ORCID to comment.