Think Before you Write: QA-Guided Reasoning for Character Descriptions in Books

TL;DR

This paper introduces a QA-guided reasoning framework for generating accurate character descriptions from long narratives, improving faithfulness and grounding by decoupling reasoning from generation.

Contribution

It proposes a novel training approach that separates reasoning and generation, enhancing character description quality in narrative applications.

Findings

QA-guided reasoning improves faithfulness and informativeness.

Decoupling reasoning from generation outperforms baseline models.

Framework is effective on datasets BookWorm and CroSS.

Abstract

Character description generation is an important capability for narrative-focused applications such as summarization, story analysis, and character-driven simulations. However, generating accurate character descriptions from long-form narratives (e.g., novels) is challenging: models must track evolving attributes (e.g., relationships and events), integrate evidence scattered across the text, and infer implicit details. Despite the success of reasoning-enabled LLMs on many benchmarks, we find that for character description generation their performance improves when built-in reasoning is disabled (i.e., an empty reasoning trace). Motivated by this, we propose a training framework that decouples reasoning from generation. Our approach, which can be applied on top of long-context LLMs or chunk-based methods, consists of a reasoning model that produces a structured QA reasoning trace and a generation model that conditions on this trace to produce the final character description. Experiments on two datasets (BookWorm and CroSS) show that QA-guided reasoning improves faithfulness, informativeness, and grounding over strong long-context baselines.