Research

Paper

TESTING March 04, 2026

LoRA-MME: Multi-Model Ensemble of LoRA-Tuned Encoders for Code Comment Classification

Authors

Md Akib Haider, Ahsan Bulbul, Nafis Fuad Shahid, Aimaan Ahmed, Mohammad Ishrak Abedin

Abstract

Code comment classification is a critical task for automated software documentation and analysis. In the context of the NLBSE'26 Tool Competition, we present \textbf{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 \textbf{F1 Weighted score of 0.7906} and a \textbf{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.

Metadata

arXiv ID: 2603.03959
Provider: ARXIV
Primary Category: cs.SE
Published: 2026-03-04
Fetched: 2026-03-05 06:06

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.03959v1</id>\n    <title>LoRA-MME: Multi-Model Ensemble of LoRA-Tuned Encoders for Code Comment Classification</title>\n    <updated>2026-03-04T11:36:32Z</updated>\n    <link href='https://arxiv.org/abs/2603.03959v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.03959v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Code comment classification is a critical task for automated software documentation and analysis. In the context of the NLBSE'26 Tool Competition, we present \\textbf{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 \\textbf{F1 Weighted score of 0.7906} and a \\textbf{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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SE'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <published>2026-03-04T11:36:32Z</published>\n    <arxiv:primary_category term='cs.SE'/>\n    <author>\n      <name>Md Akib Haider</name>\n    </author>\n    <author>\n      <name>Ahsan Bulbul</name>\n    </author>\n    <author>\n      <name>Nafis Fuad Shahid</name>\n    </author>\n    <author>\n      <name>Aimaan Ahmed</name>\n    </author>\n    <author>\n      <name>Mohammad Ishrak Abedin</name>\n    </author>\n  </entry>"
}