Evaluating Code Comment Generation With Summarized API Docs
Bibliographic record
Abstract
Code comment generation is the task of generating a high-level natural language description for a given code snippet. API2Com is a comment generation model designed to leverage the Application Programming Interface Documentations (API Docs) as an external knowledge resource. Shahbazi et al. [1] showed that API Docs might help increase the model's performance. However, the model's performance in generating pertinent comments deteriorates due to the lengthy documentation used in the input as the number of APIs used in a method increases. In this paper, we propose to evaluate how summarizing the API Docs using an extractive text summarization technique, TextRank, will impact the overall performance of the API2Com. The results of our experiments using the same Java dataset confirm the inverse correlation between the number of APIs and the model's performance. As the number of APIs increases, the performance metrics tend to deteriorate for both configurations of the model, with or without API Docs summarization using TextRank. Experiments also show the impact of the number of APIs on TextRank algorithm capacity to improve the model per-formance. For example, with 8 APIs, TextRank summarization improved the model BLEU score by 18% on average, but the performance tends to decrease as the number of APIs increases. This demonstrates an open area of research to determine the winning combination in terms of the model configuration and the length of documentation used.
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 | 0.004 | 0.022 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| 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".