Large Language Models Can Learn Temporal Reasoning

2024/01/12 by Siheng Xiong, Ali Payani, Xiong, Siheng +5 · 29 citations
Computer Science · Social Sciences · #Computation and Language (cs.CL) #Computational and Text Analysis Methods #FOS: Computer and information sciences #Natural Language Processing Techniques #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2401.06853

openalex publication_date 2024/01/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

While large language models (LLMs) have demonstrated remarkable reasoning capabilities, they are not without their flaws and inaccuracies. Recent studies have introduced various methods to mitigate these limitations. Temporal reasoning (TR), in particular, presents a significant challenge for LLMs due to its reliance on diverse temporal concepts and intricate temporal logic. In this paper, we propose TG-LLM, a novel framework towards language-based TR. Instead of reasoning over the original context, we adopt a latent representation, temporal graph (TG) that enhances the learning of TR. A synthetic dataset (TGQA), which is fully controllable and requires minimal supervision, is constructed for fine-tuning LLMs on this text-to-TG translation task. We confirmed in experiments that the capability of TG translation learned on our dataset can be transferred to other TR tasks and benchmarks. On top of that, we teach LLM to perform deliberate reasoning over the TGs via Chain-of-Thought (CoT) bootstrapping and graph data augmentation. We observed that those strategies, which maintain a balance between usefulness and diversity, bring more reliable CoTs and final results than the vanilla CoT distillation.

Cited by

• A Stepwise-Enhanced Reasoning Framework for Large Language Models Based on External Subgraph Generation
• Accounting Reasoning in Large Language Models: Concepts, Evaluation, and Empirical Analysis
• HybridFlow: Resource-Adaptive Subtask Routing for Efficient Edge-Cloud LLM Inference
• Knowledge Reasoning of Large Language Models Integrating Graph-Structured Information for Pest and Disease Control in Tobacco
• Semantic Deception: When Reasoning Models Can't Compute an Addition
• Memory-T1: Reinforcement Learning for Temporal Reasoning in Multi-session Agents
• A Large Language Model Based Method for Complex Logical Reasoning over Knowledge Graphs
• NeSTR: A Neuro-Symbolic Abductive Framework for Temporal Reasoning in Large Language Models
• SPAN: Benchmarking and Improving Cross-Calendar Temporal Reasoning of Large Language Models
• TempoBench: Evaluating Temporal Causal Reasoning in Large Language Models
• Temporal Blindness in Multi-Turn LLM Agents: Misaligned Tool Use vs. Human Time Perception
• Temporal Referential Consistency: Do LLMs Favor Sequences Over Absolute Time References?
• A Mimamsa Inspired Framework For Instruction Sequencing In AI Agents
• See or Say Graphs: Agent-Driven Scalable Graph Understanding with Vision-Language Models
• Structured yet Bounded Temporal Understanding in Large Language Models
• Enhancing Long Chain-of-Thought Reasoning through Multi-Path Plan Aggregation
• GTR-Bench: Evaluating Geo-Temporal Reasoning in Vision-Language Models
• Enhancing Time Awareness in Generative Recommendation
• HistoryBankQA: Multilingual Temporal Question Answering on Historical Events
• A Survey of Reasoning and Agentic Systems in Time Series with Large Language Models
• Self-Exploring Language Models for Explainable Link Forecasting on Temporal Graphs via Reinforcement Learning
• ConceptBot: Enhancing Robot's Autonomy through Task Decomposition with Large Language Models and Knowledge Graph
• Analysis of Semantic Communication for Logic-based Hypothesis Deduction
• MobQA: A Benchmark Dataset for Semantic Understanding of Human Mobility Data through Question Answering
• Reasoning in Computer Vision: Taxonomy, Models, Tasks, and Methodologies
• Uncertainty-quantified Rollout Policy Adaptation for Unlabelled Cross-domain Temporal Grounding
• Harnessing Temporal Databases for Systematic Evaluation of Factual Time-Sensitive Question-Answering in Large Language Models
• T-GRAG: A Dynamic GraphRAG Framework for Resolving Temporal Conflicts and Redundancy in Knowledge Retrieval
• Who's important? -- SUnSET: Synergistic Understanding of Stakeholder, Events and Time for Timeline Generation

Related

• LTLBench: Towards Benchmarks for Evaluating Temporal Logic Reasoning in Large Language Models
• Large Language Models-guided Dynamic Adaptation for Temporal Knowledge Graph Reasoning
• Learning to Reason Over Time: Timeline Self-Reflection for Improved Temporal Reasoning in Language Models
• Narrative-of-Thought: Improving Temporal Reasoning of Large Language Models via Recounted Narratives
• Chain of History: Learning and Forecasting with LLMs for Temporal Knowledge Graph Completion