LoRA-MME: Multi-Model Ensemble of LoRA-Tuned Encoders for Code Comment Classification
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913034494017536 |
|---|---|
| author | Haider, Md Akib Bulbul, Ahsan Shahid, Nafis Fuad Ahmed, Aimaan Abedin, Mohammad Ishrak |
| author_facet | Haider, Md Akib Bulbul, Ahsan Shahid, Nafis Fuad Ahmed, Aimaan Abedin, Mohammad Ishrak |
| contents | Code comment classification is a critical task for automated software documentation and analysis. In the context of the NLBSE'26 Tool Competition, we present LoRA-MME, a Multi-Model Ensemble architecture utilizing Parameter-Efficient Fine-Tuning (PEFT). Our approach addresses the multi-label classification challenge across Java, Python, and Pharo by combining the strengths of four distinct transformer encoders: UniXcoder, CodeBERT, GraphCodeBERT, and CodeBERTa. By independently fine-tuning these models using Low-Rank Adaptation(LoRA) and aggregating their predictions via a learned weighted ensemble strategy, we maximize classification performance without the memory overhead of full model fine-tuning. Our tool achieved an F1 Weighted score of 0.7906 and a Macro F1 of 0.6867 on the test set. However, the computational cost of the ensemble resulted in a final submission score of 41.20%, highlighting the trade-off between semantic accuracy and inference efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_03959 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LoRA-MME: Multi-Model Ensemble of LoRA-Tuned Encoders for Code Comment Classification Haider, Md Akib Bulbul, Ahsan Shahid, Nafis Fuad Ahmed, Aimaan Abedin, Mohammad Ishrak Software Engineering Machine Learning Code comment classification is a critical task for automated software documentation and analysis. In the context of the NLBSE'26 Tool Competition, we present LoRA-MME, a Multi-Model Ensemble architecture utilizing Parameter-Efficient Fine-Tuning (PEFT). Our approach addresses the multi-label classification challenge across Java, Python, and Pharo by combining the strengths of four distinct transformer encoders: UniXcoder, CodeBERT, GraphCodeBERT, and CodeBERTa. By independently fine-tuning these models using Low-Rank Adaptation(LoRA) and aggregating their predictions via a learned weighted ensemble strategy, we maximize classification performance without the memory overhead of full model fine-tuning. Our tool achieved an F1 Weighted score of 0.7906 and a Macro F1 of 0.6867 on the test set. However, the computational cost of the ensemble resulted in a final submission score of 41.20%, highlighting the trade-off between semantic accuracy and inference efficiency. |
| title | LoRA-MME: Multi-Model Ensemble of LoRA-Tuned Encoders for Code Comment Classification |
| topic | Software Engineering Machine Learning |
| url | https://arxiv.org/abs/2603.03959 |