A Logically Consistent Chain-of-Thought Approach for Stance Detection
This work improves stance detection for unseen targets in NLP applications, representing an incremental advancement in knowledge-enhanced methods.
The paper tackles the problem of zero-shot stance detection (ZSSD) by addressing knowledge-task disconnect and logical inconsistency, introducing the Logically Consistent Chain-of-Thought (LC-CoT) approach, which outperforms traditional supervised methods without labeled data.
Zero-shot stance detection (ZSSD) aims to detect stances toward unseen targets. Incorporating background knowledge to enhance transferability between seen and unseen targets constitutes the primary approach of ZSSD. However, these methods often struggle with a knowledge-task disconnect and lack logical consistency in their predictions. To address these issues, we introduce a novel approach named Logically Consistent Chain-of-Thought (LC-CoT) for ZSSD, which improves stance detection by ensuring relevant and logically sound knowledge extraction. LC-CoT employs a three-step process. Initially, it assesses whether supplementary external knowledge is necessary. Subsequently, it uses API calls to retrieve this knowledge, which can be processed by a separate LLM. Finally, a manual exemplar guides the LLM to infer stance categories, using an if-then logical structure to maintain relevance and logical coherence. This structured approach to eliciting background knowledge enhances the model's capability, outperforming traditional supervised methods without relying on labeled data.