RuleFlow : Generating Reusable Program Optimizations with LLMs
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910016816021504 |
|---|---|
| author | Singh, Avaljot Bharadwaj, Dushyant Baziotis, Stefanos Varadharajan, Kaushik Mendis, Charith |
| author_facet | Singh, Avaljot Bharadwaj, Dushyant Baziotis, Stefanos Varadharajan, Kaushik Mendis, Charith |
| contents | Optimizing Pandas programs is a challenging problem. Existing systems and compiler-based approaches offer reliability but are either heavyweight or support only a limited set of optimizations. Conversely, using LLMs in a per-program optimization methodology can synthesize nontrivial optimizations, but is unreliable, expensive, and offers a low yield. In this work, we introduce a hybrid approach that works in a 3-stage manner that decouples discovery from deployment and connects them via a novel bridge. First, it discovers per-program optimizations (discovery). Second, they are converted into generalised rewrite rules (bridge). Finally, these rules are incorporated into a compiler that can automatically apply them wherever applicable, eliminating repeated reliance on LLMs (deployment). We demonstrate that RuleFlow is the new state-of-the-art (SOTA) Pandas optimization framework on PandasBench, a challenging Pandas benchmark consisting of Python notebooks. Across these notebooks, we achieve a speedup of up to 4.3x over Dias, the previous compiler-based SOTA, and 1914.9x over Modin, the previous systems-based SOTA.
Our code is available at https://github.com/ADAPT-uiuc/RuleFlow. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_09051 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RuleFlow : Generating Reusable Program Optimizations with LLMs Singh, Avaljot Bharadwaj, Dushyant Baziotis, Stefanos Varadharajan, Kaushik Mendis, Charith Software Engineering Artificial Intelligence Optimizing Pandas programs is a challenging problem. Existing systems and compiler-based approaches offer reliability but are either heavyweight or support only a limited set of optimizations. Conversely, using LLMs in a per-program optimization methodology can synthesize nontrivial optimizations, but is unreliable, expensive, and offers a low yield. In this work, we introduce a hybrid approach that works in a 3-stage manner that decouples discovery from deployment and connects them via a novel bridge. First, it discovers per-program optimizations (discovery). Second, they are converted into generalised rewrite rules (bridge). Finally, these rules are incorporated into a compiler that can automatically apply them wherever applicable, eliminating repeated reliance on LLMs (deployment). We demonstrate that RuleFlow is the new state-of-the-art (SOTA) Pandas optimization framework on PandasBench, a challenging Pandas benchmark consisting of Python notebooks. Across these notebooks, we achieve a speedup of up to 4.3x over Dias, the previous compiler-based SOTA, and 1914.9x over Modin, the previous systems-based SOTA. Our code is available at https://github.com/ADAPT-uiuc/RuleFlow. |
| title | RuleFlow : Generating Reusable Program Optimizations with LLMs |
| topic | Software Engineering Artificial Intelligence |
| url | https://arxiv.org/abs/2602.09051 |