When Reasoning Supervision Hurts: TTCW-Based Long-Form Literary Review Generation

Abstract:Automatic evaluation of long-form literary writing remains challenging, particularly for creativity-related dimensions such as originality and flexibility. We address the lack of large-scale supervision for structured Torrance Test of Creative Writing (TTCW)-based evaluation by constructing a dataset of 263{,}911 long-form stories, each paired with scalar scores and meta-synthesised review comments across 14 TTCW-based dimensions. Using this dataset, we study whether reviewer-style reasoning traces improve structured literary review generation across Qwen3-4B, Qwen3-8B, and Nemotron-8B. Across all three models, no-trace supervision yields higher score imitation fidelity and near-complete report generation, whereas trace supervision reduces completion to $85.9$--$90.9\%$. Failure analysis shows that $88.6\%$ of failed reasoning generations contain semantic repetition loops inside the trace. These loops concentrate in particular story--criterion interactions and progressively harden after onset, while comparable repetition is rare in the supervision data. Our results suggest that, for long-form tasks requiring strictly structured outputs, explicit reasoning traces can introduce generation instability that outweighs their overall benefit.

Computation and Language (cs.CL)

arXiv:2605.20364 [cs.CL]

(or arXiv:2605.20364v2 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2605.20364 Focus to learn more arXiv-issued DOI via DataCite

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.