AnchorAL is a computationally efficient active learning method that addresses class imbalance by dynamically selecting a small, balanced subpool of instances to run the active learning strategy on, promoting the discovery of minority instances.
Incorporating human-annotated rationales into text classification models enhances the plausibility of post-hoc explanations without substantially degrading model performance.
Even minor changes to prompts can significantly alter the predictions of large language models, with some variations leading to substantial performance degradation.
텍스트 분류를 위한 RulePrompt 방법은 PLM과 논리 규칙을 활용하여 효과적으로 분류 성능을 향상시킵니다.
Die Mutual Reinforcement Effect (MRE) Theorie wird durch Information Flow Analyse in Textklassifikationsaufgaben bestätigt.
TELEClass verbessert die hierarchische Textklassifizierung durch Taxonomieanreicherung und LLM-Verbesserung.
Hierarchical text classification with minimal supervision using Taxonomy Enrichment and LLM enhancement.
Hierarchical text classification with minimal supervision using taxonomy enrichment and LLM enhancement.
The authors propose RulePrompt, a novel approach that leverages logical rules to enhance the understanding of categories in weakly supervised text classification tasks using prompting PLMs.
The author analyzes the Margin Discrepancy-based Adversarial Training approach for Multi-Domain Text Classification, providing theoretical underpinnings and empirical validation.