Towards Explainable Temporal Reasoning in Large Language Models: A Structure-Aware Generative Framework

2025/05/21 by Zihao Jiang, Ben Liu, Jiang, Zihao +11 · 1 citation
Computer Science · Decision Sciences · #Artificial Intelligence (cs.AI) #Big Data Technologies and Applications #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2505.15245

openalex publication_date 2025/05/21 · openalex created_date 2025/10/19 · openalex updated_date 2026/07/28

Abstract

While large language models (LLMs) show great potential in temporal reasoning, most existing work focuses heavily on enhancing performance, often neglecting the explainable reasoning processes underlying the results. To address this gap, we introduce a comprehensive benchmark covering a wide range of temporal granularities, designed to systematically evaluate LLMs' capabilities in explainable temporal reasoning. Furthermore, our findings reveal that LLMs struggle to deliver convincing explanations when relying solely on textual information. To address challenge, we propose GETER, a novel structure-aware generative framework that integrates Graph structures with text for Explainable TEmporal Reasoning. Specifically, we first leverage temporal knowledge graphs to develop a temporal encoder that captures structural information for the query. Subsequently, we introduce a structure-text prefix adapter to map graph structure features into the text embedding space. Finally, LLMs generate explanation text by seamlessly integrating the soft graph token with instruction-tuning prompt tokens. Experimental results indicate that GETER achieves state-of-the-art performance while also demonstrating its effectiveness as well as strong generalization capabilities. Our dataset and code are available at https://github.com/carryTatum/GETER.

Citations

• Rewarding Graph Reasoning Process makes LLMs more Generalized Reasoners
• Unfolding the Headline: Iterative Self-Questioning for News Retrieval and Timeline Summarization
• Enhancing Temporal Sensitivity and Reasoning for Time-Sensitive Question Answering
• XRec: Large Language Models for Explainable Recommendation
• Chain-of-History Reasoning for Temporal Knowledge Graph Forecasting
• Large Language Models Can Learn Temporal Reasoning
• ReasoningLM: Enabling Structural Subgraph Reasoning in Pre-trained Language Models for Question Answering over Knowledge Graph
• Towards Robust Temporal Reasoning of Large Language Models via a Multi-Hop QA Dataset and Pseudo-Instruction Tuning
• Temporal Knowledge Graph Forecasting Without Knowledge Using In-Context Learning
• Temporal Knowledge Graph Reasoning with Historical Contrastive Learning
• Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning
• Self-Consistency Improves Chain of Thought Reasoning in Language Models
• DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing
• TimeTraveler: Reinforcement Learning for Temporal Knowledge Graph Forecasting
• Temporal Knowledge Graph Reasoning Based on Evolutional Representation Learning
• ZeRO: Memory Optimizations Toward Training Trillion Parameter Models
• Learning Sequence Encoders for Temporal Knowledge Graph Completion

Cited by

• PredAct-Bench: Benchmarking Tool-Augmented Dialogue under Controlled Tool Noise

Related

• Large Language Models Can Learn Temporal Reasoning
• Back to the Future: Towards Explainable Temporal Reasoning with Large Language Models
• MemoTime: Memory-Augmented Temporal Knowledge Graph Enhanced Large Language Model Reasoning
• Temporal Reasoning with Large Language Models Augmented by Evolving Knowledge Graphs
• Large Language Models-guided Dynamic Adaptation for Temporal Knowledge Graph Reasoning