Pith. sign in

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

arxiv 2404.18864 v1 pith:NS72SBDD submitted 2024-04-29 cs.DC cs.AIcs.SE

classification cs.DCcs.AIcs.SE
keywords codellmsperformancedifficultlargemodelmodelssoftware
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
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.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Effect of Abstractions and Prompting Strategies on LLM-Guided High-Performance Optimizations

    cs.DC 2026-08 conditional novelty 6.0 of 10

    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.

  2. HPC-Coder-V2: Studying Code LLMs Across Low-Resource Parallel Languages

    cs.DC 2024-12 conditional novelty 6.0 of 10

    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.

  3. Can Large Language Models Predict Parallel Code Performance?

    cs.DC 2025-05 conditional novelty 5.0 of 10

    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.

  4. Language Models for Code Optimization: Survey, Challenges and Future Directions

    cs.SE 2025-01 accept novelty 5.0 of 10

    A systematic review of 53 papers on using large language models for code optimization, with a taxonomy, five challenges, and eight future research directions.

Pith tools