Papers by Taegyoon Kim
Is External Information Useful for Stance Detection with LLMs? (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing studies suggest that external information, such as Wikipedia excerpts, can improve stance detection performance, but whether such information can benefit large language models remains an unanswered question. |
| Approach: | They conduct a systematic evaluation on how Wikipedia and web search external information can affect stance detection across eight LLMs and in three datasets with 12 targets. |
| Outcome: | The proposed model degrades performance in most cases, with macro F1 scores dropping by up to 27.9%, while fine-tuning mitigates but does not eliminate it. |