analiza aiqbench
This research addresses the inefficiency of large language models when editing code originally generated by different models, noting a tendency toward excessive stylistic changes. The proposed CROCODIL framework utilizes a dual-reward system to minimize unnecessary edits while maintaining functional correctness across mixed-model workflows. This advancement is significant for optimizing collaborative development environments where multiple AI assistants are routinely used.
CROCODIL: Cross-Model Code Editing with LLMs
arXiv:2609.03894v1 Announce Type: new Abstract: Large language models (LLMs) have become ubiquitous tools for code generation and editing. However, development teams often use multiple LLM assistants. Different developers may prefer different models, and individual developers may switch between models across different coding sessions. Because of this, the edits any one model makes are frequently applied to foreign code originally generated by another model. These LLMs are often trained on different datasets, and as a result have different stylistic preferences. Do LLMs behave differently when they edit foreign code originally written by a different LLM with a different coding style? We find that models tend to make more, and often excessive, edits on foreign code. We introduce CROCODIL (Cross-model Code Editing with LLMs), a post-training framework for reducing excessive edits while preserving functional correctness. CROCODIL's similarity reward penalizes large changes, while its execution reward scores build and test success. We use the product of these two rewards to encourage the policy to decrease the edit size without decreasing the edit task success rate. CROCODIL is available at https://github.com/EngineeringSoftware/Crocodil.
Abstract:Large language models (LLMs) have become ubiquitous tools for code generation and editing. However, development teams often use multiple LLM assistants. Different developers may prefer different models, and individual developers may switch between models across different coding sessions. Because of this, the edits any one model makes are frequently applied to foreign code originally generated by another model. These LLMs are often trained on different datasets, and as a result have different stylistic preferences. Do LLMs behave differently when they edit foreign code originally written by a different LLM with a different coding style? We find that models tend to make more, and often excessive, edits on foreign code. We introduce CROCODIL (Cross-model Code Editing with LLMs), a post-training framework for reducing excessive edits while preserving functional correctness. CROCODIL's similarity reward penalizes large changes, while its execution reward scores build and test success. We use the product of these two rewards to encourage the policy to decrease the edit size without decreasing the edit task success rate. CROCODIL is available at this https URL.
Computation and Language (cs.CL); Software Engineering (cs.SE)
arXiv:2609.03894 [cs.CL]
(or arXiv:2609.03894v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2609.03894 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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.
Text integral sub licența deschisă a publicației · 272 cuvinte
Subiecte conexe
The Continual Constitutional Convention: How language models revise the documents that govern them
AI Alignment Forum · 13 oct. 2026
[AINews] The Future of Latent Space
Latent Space · 25 sept. 2026
Links #3: 2026/06 Part 1
LessWrong · 15 iun. 2026
[AINews] Opus 5.5 is good at explainer videos
Latent Space · 29 sept. 2026