SymbolicThought: Integrating Language Models and Symbolic Reasoning for Consistent and Interpretable Human Relationship Understanding
This addresses the need for consistent and interpretable human relationship understanding in narrative analysis and socially grounded AI research, offering a practical tool for tasks like explainable AI and LLM evaluation.
The paper tackles the problem of understanding character relationships in narratives by introducing SymbolicThought, a framework that integrates language models with symbolic reasoning to improve accuracy and consistency, reducing annotation time while providing interpretable outputs.
Understanding character relationships is essential for interpreting complex narratives and conducting socially grounded AI research. However, manual annotation is time-consuming and low in coverage, while large language models (LLMs) often produce hallucinated or logically inconsistent outputs. We present SymbolicThought, a human-in-the-loop framework that combines LLM-based extraction with symbolic reasoning. The system constructs editable character relationship graphs, refines them using seven types of logical constraints, and enables real-time validation and conflict resolution through an interactive interface. To support logical supervision and explainable social analysis, we release a dataset of 160 interpersonal relationships with corresponding logical structures. Experiments show that SymbolicThought improves annotation accuracy and consistency while significantly reducing time cost, offering a practical tool for narrative understanding, explainable AI, and LLM evaluation.