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.

翻译：代码注释分类是自动化软件文档与分析中的关键任务。针对NLBSE'26工具竞赛，本文提出LoRA-MME——一种利用参数高效微调技术的多模型集成架构。该方法通过整合四种不同Transformer编码器（UniXcoder、CodeBERT、GraphCodeBERT与CodeBERTa）的优势，应对Java、Python和Pharo语言的多标签分类挑战。通过使用低秩自适应技术独立微调各模型，并采用可学习的加权集成策略聚合其预测结果，我们在避免全模型微调内存开销的同时最大化分类性能。我们的工具在测试集上取得了0.7906的加权F1分数和0.6867的宏平均F1分数。然而，集成方法带来的计算成本导致最终提交分数为41.20%，这凸显了语义准确性与推理效率之间的权衡关系。