Saved in:
Bibliographic Details
Main Authors: Li, Zhaoyi, Zhang, Xu, Wan, Xiaojun
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.14410
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915932990865408
author Li, Zhaoyi
Zhang, Xu
Wan, Xiaojun
author_facet Li, Zhaoyi
Zhang, Xu
Wan, Xiaojun
contents Generating long-form linear fiction from open-ended themes remains a major challenge for large language models, which frequently fail to guarantee global structure and narrative diversity when using premise-based or linear outlining approaches. We present BiT-MCTS, a theme-driven framework that operationalizes a "climax-first, bidirectional expansion" strategy motivated by Freytag's Pyramid. Given a theme, our method extracts a core dramatic conflict and generates an explicit climax, then employs a bidirectional Monte Carlo Tree Search (MCTS) to expand the plot backward (rising action, exposition) and forward (falling action, resolution) to produce a structured outline. A final generation stage realizes a complete narrative from the refined outline. We construct a Chinese theme corpus for evaluation and conduct extensive experiments across three contemporary LLM backbones. Results show that BiT-MCTS improves narrative coherence, plot structure, and thematic depth relative to strong baselines, while enabling substantially longer, more coherent stories according to automatic metrics and human judgments.
format Preprint
id arxiv_https___arxiv_org_abs_2603_14410
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle BiT-MCTS: A Theme-based Bidirectional MCTS Approach to Chinese Fiction Generation
Li, Zhaoyi
Zhang, Xu
Wan, Xiaojun
Computation and Language
Generating long-form linear fiction from open-ended themes remains a major challenge for large language models, which frequently fail to guarantee global structure and narrative diversity when using premise-based or linear outlining approaches. We present BiT-MCTS, a theme-driven framework that operationalizes a "climax-first, bidirectional expansion" strategy motivated by Freytag's Pyramid. Given a theme, our method extracts a core dramatic conflict and generates an explicit climax, then employs a bidirectional Monte Carlo Tree Search (MCTS) to expand the plot backward (rising action, exposition) and forward (falling action, resolution) to produce a structured outline. A final generation stage realizes a complete narrative from the refined outline. We construct a Chinese theme corpus for evaluation and conduct extensive experiments across three contemporary LLM backbones. Results show that BiT-MCTS improves narrative coherence, plot structure, and thematic depth relative to strong baselines, while enabling substantially longer, more coherent stories according to automatic metrics and human judgments.
title BiT-MCTS: A Theme-based Bidirectional MCTS Approach to Chinese Fiction Generation
topic Computation and Language
url https://arxiv.org/abs/2603.14410