MLGN:A multi-label guided network for improving text classification

Within natural language processing, multi-label classification is an important but challenging task. It more complex than single-label since the document representations need to cover fine-grained label information, while labels predicted by model are often related. Recently, large pre-trained models have achieved great performance on tasks, typically using embedding of [CLS] vector as semantic representation entire and matching it with candidate labels. However, existing methods tend ignore semantics, relationships between documents not effectively mined. In addition, linear layers used for fine-tuning do take correlations into account. this work, we propose a Multi-Label Guided Network (MLGN) capable guide information. Furthermore, utilize correlation knowledge enhance original prediction in downstream tasks. The extensive experimental trials show that MLGN transcends previous works several publicly available datasets. Our source code at https://github.com/L199Q/MLGN.