Test transfer across languages and narrative traditions

Determine whether the seven-shape taxonomy, structural evaluation bands, and model results transfer to narrative traditions and languages other than the English-language tellings used for annotation.

Background

All event descriptions, graph annotations, and anonymized inputs are written in English, even though the corpus includes works originating in several other languages and traditions. The corpus therefore evaluates English-language representations of these narratives rather than multilingual narrative structures directly.

The taxonomy, structural bands, and model findings may not generalize to other languages’ narrative conventions, but this has not been tested.

References

The works themselves are not all Anglophone, and the corpus includes works first told in Korean, Japanese, Russian, German, French and Spanish. Each is annotated from an English-language telling or translation, however, and whether the taxonomy, the bands or the results transfer to other languages' narrative traditions is untested.

— Can LLMs Tell Stories Out of Order? A Corpus and Benchmark for Nonlinear Narrative Generation  (2610.21468 - Raiyan et al., 15 Oct 2026) in Limitations, paragraph “English only”