Living the Novel: A System for Generating Self-Training Timeline-Aware Conversational Agents from Novels

2025-12-08

AI Summary

This paper addresses two key challenges in literary immersive dialogue with large language models: persona drift and narrative logic breakdown (e.g., spoilers, frame violations). To tackle these, we propose a two-stage training framework that innovatively integrates data-free reinforcement learning fine-tuning (RLFT), retrieval-augmented training, and a story timeline-aware knowledge graph—enabling deep persona alignment and explicit narrative constraint modeling. Our method incorporates persona self-training and interference-resilient mechanisms, significantly improving stability and consistency in multi-character mobile dialogues. Empirical evaluation on *Twenty Thousand Leagues Under the Sea* demonstrates superior persona consistency over GPT-4o and achieves >98% narrative coherence and robustness. User diary analysis further confirms long-term interaction reliability. The core contribution is the first integration of temporal-aware knowledge graphs with unsupervised persona alignment, delivering a scalable solution for narrative consistency in literary AI agents.

Technology Category

• Natural Language Processing: Conversational AI/Dialog Systems
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

We present the Living Novel, an end-to-end system that transforms any literary work into an immersive, multi-character conversational experience. This system is designed to solve two fundamental challenges for LLM-driven characters. Firstly, generic LLMs suffer from persona drift, often failing to stay in character. Secondly, agents often exhibit abilities that extend beyond the constraints of the story's world and logic, leading to both narrative incoherence (spoiler leakage) and robustness failures (frame-breaking). To address these challenges, we introduce a novel two-stage training pipeline. Our Deep Persona Alignment (DPA) stage uses data-free reinforcement finetuning to instill deep character fidelity. Our Coherence and Robustness Enhancing (CRE) stage then employs a story-time-aware knowledge graph and a second retrieval-grounded training pass to architecturally enforce these narrative constraints. We validate our system through a multi-phase evaluation using Jules Verne's Twenty Thousand Leagues Under the Sea. A lab study with a detailed ablation of system components is followed by a 5-day in-the-wild diary study. Our DPA pipeline helps our specialized model outperform GPT-4o on persona-specific metrics, and our CRE stage achieves near-perfect performance in coherence and robustness measures. Our study surfaces practical design guidelines for AI-driven narrative systems: we find that character-first self-training is foundational for believability, while explicit story-time constraints are crucial for sustaining coherent, interruption-resilient mobile-web experiences.

Problem

• Addresses persona drift in LLM-driven characters for narrative fidelity.
• Prevents spoiler leakage and frame-breaking to maintain story coherence.
• Enforces narrative constraints using timeline-aware knowledge graphs.

Innovation

• Two-stage training pipeline for character fidelity and narrative coherence
• Deep Persona Alignment using data-free reinforcement finetuning
• Coherence and Robustness Enhancing with story-time-aware knowledge graph