LLM-Based Embeddings for Program Analysis and Optimization
Mirrored from arXiv — Machine Learning for archival readability. Support the source by reading on the original site.
Computer Science > Machine Learning
Title:LLM-Based Embeddings for Program Analysis and Optimization
Abstract:Recent advances have highlighted the potential of machine learning, particularly Large Language Models (LLMs), for analyzing and optimizing programs. We present the first application of program embeddings from LLMCompiler---an LLM massively pretrained on intermediate representation (IR) code---to representative program analysis and optimization tasks. We generate program embeddings directly from source and IR code using a simple approach: split programs into chunks, independently embed each chunk with pretrained LLMs, and then aggregate the chunk embeddings into a single program embedding. Our experiments show that combining source and IR code embeddings achieves an error rate of 1.54\% in algorithm classification, a 12\% improvement over the current state-of-the-art, and a competitive accuracy on heterogeneous device mapping. These findings suggest that training a performance-aware LLM for embedding IR code might yield state-of-the-art results in code optimization tasks.
| Subjects: | Machine Learning (cs.LG); Programming Languages (cs.PL) |
| Cite as: | arXiv:2608.07894 [cs.LG] |
| (or arXiv:2608.07894v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2608.07894
arXiv-issued DOI via DataCite (pending registration)
|
|
| Journal reference: | 2025 International Joint Conference on Neural Networks (IJCNN), Rome, Italy, 2025, pp. 1-8 |
| Related DOI: | https://doi.org/10.1109/IJCNN64981.2025.11227194
DOI(s) linking to related resources
|
Access Paper:
- View PDF
- HTML (experimental)
- TeX Source
References & Citations
Bookmark
Bibliographic and Citation Tools
Code, Data and Media Associated with this Article
Demos
Recommenders and Search Tools
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
More from arXiv — Machine Learning
-
LoKiFormer: Locality-aware Attention with Decoupled Knowledge Memory for Efficient Large Language Model Pretraining
Aug 14
-
Which Site, and When: A Free-Satellite-Data Test of Himalayan Glacial Lake Bursts, Landslides, and Ice Floods
Aug 14
-
MARCH: Scaling Recurrent Memory with Content-Routed State Anchors
Aug 14
-
Multi-AUV Ad-hoc network-based Target Tracking: A Value Gradient Guidance Multi-Agent Diffusion Reinforcement Learning Approach
Aug 14
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.