GPTCloneBench: A comprehensive benchmark of semantic clones and cross-language clones using GPT-3 model and SemanticCloneBench
Bibliographic record
Abstract
With the emergence of Machine Learning, there has been a surge in leveraging its capabilities for problem-solving across various domains. In the code clone realm, the identification of type-4 or semantic clones has emerged as a crucial yet challenging task. Researchers aim to utilize Machine Learning to tackle this challenge, often relying on the Big-CloneBench dataset. However, it’s worth noting that BigCloneBench, originally not designed for semantic clone detection, presents several limitations that hinder its suitability as a comprehensive training dataset for this specific purpose. Furthermore, CLCDSA dataset suffers from a lack of reusable examples aligning with real-world software systems, rendering it inadequate for cross-language clone detection approaches. In this work, we present a comprehensive semantic clone and cross-language clone benchmark, GPTCloneBench1by exploiting SemanticCloneBench and OpenAI’s GPT-3 model. In particular, using code fragments from SemanticCloneBench as sample inputs along with appropriate prompt engineering for GPT-3 model, we generate semantic and cross-language clones for these specific fragments and then conduct a combination of extensive manual analysis, tool-assisted filtering, functionality testing and automated validation in building the benchmark. From 79,928 clone pairs of GPT-3 output, we created a benchmark with 37,149 true semantic clone pairs, 19,288 false semantic pairs(Type-1/Type-2), and 20,770 cross-language clones across four languages (Java, C, C#, and Python). Our benchmark is 15-fold larger than SemanticCloneBench, has more functional code examples for software systems and programming language support than CLCDSA, and overcomes BigCloneBench’s qualities, quantification, and language variety limitations. GPTCloneBench can be found here1.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.016 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.006 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.005 | 0.003 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".