Narrative Studio：使用 LLM 和蒙特卡洛树搜索的视觉叙事探索
Narrative Studio: Visual narrative exploration using LLMs and Monte Carlo Tree Search
AI总结：
针对当前 LLM 聊天界面限制交互式叙事为单一流程的问题，提出 Narrative Studio，通过树状界面、迭代 LLM 推理、MCTS 自动扩展及实体图锚定，实现分支叙事的探索与增强。
摘要：
交互式叙事得益于规划和探索多种“如果”场景。现代 LLM 是构思和探索的有用工具，但当前基于聊天的用户界面将用户限制在单一的线性流程中。为了解决这一限制，我们提出了 Narrative Studio——一种新颖的浏览器内叙事探索环境，具有树状界面，允许从故事中用户定义的点进行分支探索。每个分支通过由系统和用户定义的提示引导的迭代 LLM 推理进行扩展。此外，我们采用蒙特卡洛树搜索（MCTS）根据用户指定的标准自动扩展有前景的叙事路径，从而实现更多样化和稳健的故事发展。我们还允许用户通过将生成的文本锚定在表示故事参与者和环境的实体图上来增强叙事连贯性。
英文摘要：
Interactive storytelling benefits from planning and exploring multiple 'what if' scenarios. Modern LLMs are useful tools for ideation and exploration, but current chat-based user interfaces restrict users to a single linear flow. To address this limitation, we propose Narrative Studio -- a novel in-browser narrative exploration environment featuring a tree-like interface that allows branching exploration from user-defined points in a story. Each branch is extended via iterative LLM inference guided by system and user-defined prompts. Additionally, we employ Monte Carlo Tree Search (MCTS) to automatically expand promising narrative paths based on user-specified criteria, enabling more diverse and robust story development. We also allow users to enhance narrative coherence by grounding the generated text in an entity graph that represents the actors and environment of the story.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
