The development of fact-checking systems for verifying the factuality of text generated by large language models (LLMs) has been advancing.In the verdict prediction step of such systems, the system determines whether claims in the generated text are supported by retrieved evidence, formulated as a natural language inference (NLI) task.This study extends the label set for verdict prediction to capture claim-evidence relationships that humans would commonly interpret as supported or refuted, even in the absence of strict logical entailment or contradiction.It also constructs a Japanese dataset comprising 28,147 instances from two sources based on this extended label set.We analyze the causes of annotation disagreement and find that ambiguity in the boundary of acceptable inference, interpretive characteristics of negative cases, and incomplete information in the evidence affect annotation variability.Using this dataset, we evaluate the performance of prompt-based verdict prediction methods and show that prompts that explicitly elicit chain-of-thought reasoning improve F1 by 4 percentage points compared to baseline.

We develop an embedding model specifically designed for Waka poetry and use it to build a model for detecting Honkadori. Waka is a tradi-tional form of old Japanese poetry that has been composed since ancient times. Honkadori is a sophisticated poetic technique in Japanese clas-sical literature where poets incorporate words or poetic sentiments from old Wakas (Honka) into their own work. First, we fine-tune a pre-trained language model using contrastive learn-ing to construct a Waka-specialized embedding model. Then, using the embedding vectors ob-tained from this model and features extracted from them, we train a machine learning model to detect the Honka (original poem) of Wakas that employ the Honkadori technique. Using paired data of Honka and Wakas that are consid-ered to use Honkadori, we evaluated the Honka detection model and demonstrated that it can detect Honka with reasonable accuracy.

To evaluate the creativity of large language models (LLMs) in Japanese, we construct three benchmarks: Japanese Creativity Questions (JCQ), Divergent Association Task (DAT), and Story Alteration Task (SAT). JCQ comprehensively evaluates creativity using LLMs. Meanwhile, DAT and SAT measure specific aspects of creative ability using embeddings. We also analyze correlations between JCQ and DAT, JCQ and SAT, and DAT and SAT. While JCQ provides comprehensive evaluation, it is relatively time and resource intensive. In contrast, DAT and SAT offer lower comprehensiveness but enable quick, low-cost assessment. Additionally, we investigate whether training with DAT contributes to enhancing LLM creativity.

We investigate the impact of politeness levels in prompts on the performance of large language models (LLMs). Polite language in human communications often garners more compliance and effectiveness, while rudeness can cause aversion, impacting response quality. We consider that LLMs mirror human communication traits, suggesting they align with human cultural norms. We assess the impact of politeness in prompts on LLMs across English, Chinese, and Japanese tasks. We observed that impolite prompts often result in poor performance, but overly polite language does not guarantee better outcomes. The best politeness level is different according to the language. This phenomenon suggests that LLMs not only reflect human behavior but are also influenced by language, particularly in different cultural contexts. Our findings highlight the need to factor in politeness for cross-cultural natural language processing and LLM usage.