A Novel Refactoring and Semantic Aware Abstract Syntax Tree Differencing Tool and a Benchmark for Evaluating the Accuracy of Diff Tools
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Software undergoes constant changes to support new requirements, address bugs, enhance performance, and ensure maintainability. Thus, developers spend a great portion of their workday trying to understand and review the code changes of their teammates. Abstract Syntax Tree (AST) diff tools were developed to overcome the limitations of line-based diff tools, which are used by the majority of developers. Despite the notable improvements brought by AST diff tools in understanding complex changes, they still suffer from serious limitations, such as (1) lacking multi-mapping support, (2) matching semantically incompatible AST nodes, (3) ignoring language clues to guide the matching process, (4) lacking refactoring awareness, and (5) lacking commit-level diff support. We propose a novel AST diff tool based on RefactoringMiner that resolves all aforementioned limitations. First, we improved RefactoringMiner to increase its statement mapping accuracy, and then we developed an algorithm that generates AST diff for a given commit or pull request based on the refactoring instances and pairs of matched program element declarations provided by RefactoringMiner. To evaluate the accuracy of our tool and compare it with the state-of-the-art tools, we created the first benchmark of AST node mappings, including 800 bug-fixing commits and 188 refactoring commits. Our evaluation showed that our tool achieved a considerably higher precision and recall, especially for refactoring commits, with an execution time that is comparable with that of the faster tools.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.009 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it