# Towards Explainable Temporal Reasoning in Large Language Models: A Structure-Aware Generative Framework Source: [Towards Explainable Temporal Reasoning in Large Language Models: A Structure-Aware Generative Framework]() Collection index: [llms.txt]() Data updated: 2026-01-04T04:22:54.530Z Authors: Zihao Jiang, Ben Liu, Miao Peng, Wenjie Xu, Yao Xiao, Zhenyan Shan, Min Peng Published: 2025-05-21T08:20:35.000Z ## 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. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Explanation Generation on ICEWS14 (test)](): Explanation Generation; ICEWS14 (test); BLEU-4: 40.54. - [Explanation Generation on GDELT (test)](): Explanation Generation; GDELT (test); BLEU-4: 34.46. - [Explanation Generation on ICEWS05-15 (test)](): Explanation Generation; ICEWS05-15 (test); BLEU-4: 45.98. - [Semantic Similarity on ICEWS18 (test)](): Semantic Similarity; ICEWS18 (test); BLEU-4: 40.39. - [Semantic Similarity on WIKI (test)](): Semantic Similarity; Wiki (test); BLEU-4: 55.52. - [Temporal Reasoning on ICEWS14 (test)](): Temporal Reasoning; ICEWS14 (test); Positive Score: 77.45. - [Temporal Reasoning on GDELT (test)](): Temporal Reasoning; GDELT (test); Positive Accuracy: 63.77. - [Temporal Reasoning on ICEWS05-15 (test)](): Temporal Reasoning; ICEWS05-15 (test); Positive Score: 78.94. - [Temporal Reasoning Prediction on ICEWS18 (test)](): Temporal Reasoning Prediction; ICEWS18 (test); Positive Accuracy: 75.78. - [Temporal Reasoning Prediction on WIKI (test)](): Temporal Reasoning Prediction; Wiki (test); Positive Performance: 99.28. ## Original sources and implementation [Original paper]() [PDF]() [Implementation]()