TKG-Thinker: Towards Dynamic Reasoning over Temporal Knowledge Graphs via Agentic Reinforcement Learning

• URL: http://arxiv.org/abs/2602.05818v2
• Date: Sat, 07 Feb 2026 04:47:42 GMT
• Title: TKG-Thinker: Towards Dynamic Reasoning over Temporal Knowledge Graphs via Agentic Reinforcement Learning
• Authors: Zihao Jiang, Miao Peng, Zhenyan Shan, Wenjie Xu, Ben Liu, Gong Chen, Ziqi Gao, Min Peng,
• Abstract summary: Temporal knowledge graph question answering (TKGQA) aims to answer time-sensitive questions by leveraging temporal knowledge bases.<n>Current prompting strategies constrain their efficacy in two primary ways.<n>We propose textbfTKG-Thinker, a novel agent equipped with autonomous planning and adaptive retrieval capabilities.
• Score: 22.089705008812217
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Temporal knowledge graph question answering (TKGQA) aims to answer time-sensitive questions by leveraging temporal knowledge bases. While Large Language Models (LLMs) demonstrate significant potential in TKGQA, current prompting strategies constrain their efficacy in two primary ways. First, they are prone to reasoning hallucinations under complex temporal constraints. Second, static prompting limits model autonomy and generalization, as it lack optimization through dynamic interaction with temporal knowledge graphs (TKGs) environments. To address these limitations, we propose \textbf{TKG-Thinker}, a novel agent equipped with autonomous planning and adaptive retrieval capabilities for reasoning over TKGs. Specifically, TKG-Thinker performs in-depth temporal reasoning through dynamic multi-turn interactions with TKGs via a dual-training strategy. We first apply Supervised Fine-Tuning (SFT) with chain of thought data to instill core planning capabilities, followed by a Reinforcement Learning (RL) stage that leverages multi-dimensional rewards to refine reasoning policies under intricate temporal constraints. Experimental results on benchmark datasets with three open-source LLMs show that TKG-Thinker achieves state-of-the-art performance and exhibits strong generalization across complex TKGQA settings.

Related papers

• LLM-Guided Knowledge Distillation for Temporal Knowledge Graph Reasoning [8.96967435213864]
  We propose an LLM-assisted distillation framework specifically designed for temporal knowledge graph reasoning.<n>The proposed approach consistently improves link prediction performance over strong distillation baselines.<n>The results highlight the potential of large language models as effective teachers for transferring temporal reasoning capability to resource-efficient TKG systems.
  arXiv  Detail & Related papers  (2026-02-16T03:27:50Z)
• Temp-R1: A Unified Autonomous Agent for Complex Temporal KGQA via Reverse Curriculum Reinforcement Learning [51.79753403262177]
  Temporal Knowledge Graph Question Answering (TKGQA) is inherently challenging, as it requires sophisticated reasoning over dynamic facts with multi-hop dependencies and complex temporal constraints.<n>We propose Temp-R1, the first autonomous end-to-end agent for TKGQA trained through reinforcement learning.<n>Our 8B- parameter Temp-R1 achieves state-of-the-art performance on MultiTQ and TimelineKGQA, improving 19.8% over strong baselines on complex questions.
  arXiv  Detail & Related papers  (2026-01-26T09:23:53Z)
• CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphs [53.199517625701475]
  CoG is a training-free framework inspired by Dual-Process Theory that mimics the interplay between intuition and deliberation.<n>CoG significantly outperforms state-of-the-art approaches in both accuracy and efficiency.
  arXiv  Detail & Related papers  (2026-01-16T07:27:40Z)
• Toward Better Temporal Structures for Geopolitical Events Forecasting [13.690434458053765]
  We study a generalization of HTKGs, Hyper-Relational Temporal Knowledge Generalized Hypergraphs (HTKGHs)<n>We first derive a formalization for HTKGHs, demonstrating their backward compatibility while supporting two complex types of facts commonly found in geopolitical incidents.<n>We then introduce the htkgh-polecat dataset, built upon the global event database POLECAT.
  arXiv  Detail & Related papers  (2026-01-01T18:45:07Z)
• Modeling Uncertainty Trends for Timely Retrieval in Dynamic RAG [35.96258615258145]
  We introduce Entropy-Trend Constraint (ETC), a training-free method that determines optimal retrieval timing by modeling the dynamics of token-level uncertainty.<n>ETC consistently outperforms strong baselines while reducing retrieval frequency.<n>It is plug-and-play, model-agnostic, and readily integrable into existing decoding pipelines.
  arXiv  Detail & Related papers  (2025-11-13T05:28:02Z)
• Plan of Knowledge: Retrieval-Augmented Large Language Models for Temporal Knowledge Graph Question Answering [23.330273675675897]
  Temporal Knowledge Graph Question Answering (TKGQA) aims to answer time-sensitive questions by leveraging factual information from Temporal Knowledge Graphs (TKGs)
  arXiv  Detail & Related papers  (2025-11-06T05:24:14Z)
• Plan Then Retrieve: Reinforcement Learning-Guided Complex Reasoning over Knowledge Graphs [52.16166558205338]
  Graph-RFT is a novel two-stage reinforcement fine-tuning KGQA framework with a 'plan-KGsearch-and-Websearch-during-think' paradigm.<n>It enables LLMs to perform autonomous planning and adaptive retrieval scheduling across KG and web sources under incomplete knowledge conditions.
  arXiv  Detail & Related papers  (2025-10-23T16:04:13Z)
• Learning Efficient and Generalizable Graph Retriever for Knowledge-Graph Question Answering [75.12322966980003]
  Large Language Models (LLMs) have shown strong inductive reasoning ability across various domains.<n>Most existing RAG pipelines rely on unstructured text, limiting interpretability and structured reasoning.<n>Recent studies have explored integrating knowledge graphs with LLMs for knowledge graph question answering.<n>We propose RAPL, a novel framework for efficient and effective graph retrieval in KGQA.
  arXiv  Detail & Related papers  (2025-06-11T12:03:52Z)
• STEP: Enhancing Video-LLMs' Compositional Reasoning by Spatio-Temporal Graph-guided Self-Training [87.58996020705258]
  Video Large Language Models (Video-LLMs) have recently shown strong derivation in basic video understanding tasks.<n>Video-LLMs struggle with compositional reasoning that requires multi-step explicit-temporal inference across object relations, interactions and events.<n>We propose STEP, a novel graph-guided self-training method that enables VideoLLMs to generate reasoning-rich finetuning data from any raw videos to improve itself.
  arXiv  Detail & Related papers  (2024-11-29T11:54:55Z)
• Large Language Models-guided Dynamic Adaptation for Temporal Knowledge Graph Reasoning [87.10396098919013]
  Large Language Models (LLMs) have demonstrated extensive knowledge and remarkable proficiency in temporal reasoning.<n>We propose a Large Language Models-guided Dynamic Adaptation (LLM-DA) method for reasoning on Temporal Knowledge Graphs.<n>LLM-DA harnesses the capabilities of LLMs to analyze historical data and extract temporal logical rules.
  arXiv  Detail & Related papers  (2024-05-23T04:54:37Z)
• GenTKG: Generative Forecasting on Temporal Knowledge Graph with Large Language Models [35.594662986581746]
  Large language models (LLMs) have ignited interest in the temporal knowledge graph (tKG) domain, where conventional embedding-based and rule-based methods dominate. We propose a novel retrieval-augmented generation framework named GenTKG combining a temporal logical rule-based retrieval strategy and few-shot parameter-efficient instruction tuning. Experiments have shown that GenTKG outperforms conventional methods of temporal relational forecasting with low computation resources.
  arXiv  Detail & Related papers  (2023-10-11T18:27:12Z)
• T-GAP: Learning to Walk across Time for Temporal Knowledge Graph Completion [13.209193437124881]
  Temporal knowledge graphs (TKGs) inherently reflect the transient nature of real-world knowledge, as opposed to static knowledge graphs. We propose T-GAP, a novel model for TKG completion that maximally utilizes both temporal information and graph structure in its encoder and decoder. Our experiments demonstrate that T-GAP achieves superior performance against state-of-the-art baselines, and competently generalizes to queries with unseen timestamps.
  arXiv  Detail & Related papers  (2020-12-19T04:45:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.