Self-Exploring Language Models for Explainable Link Forecasting on Temporal Graphs via Reinforcement Learning

• URL: http://arxiv.org/abs/2509.00975v2
• Date: Mon, 13 Oct 2025 02:09:16 GMT
• Title: Self-Exploring Language Models for Explainable Link Forecasting on Temporal Graphs via Reinforcement Learning
• Authors: Zifeng Ding, Shenyang Huang, Zeyu Cao, Emma Kondrup, Zachary Yang, Xingyue Huang, Yuan Sui, Zhangdie Yuan, Yuqicheng Zhu, Xianglong Hu, Yuan He, Farimah Poursafaei, Michael Bronstein, Andreas Vlachos,
• Abstract summary: We present Reasoning-Enhanced Learning for Temporal Graphs (ReaL-TG)<n>ReaL-TG is a reinforcement learning framework that fine-tunes graph models to perform explainable link forecasting.<n>We propose a new evaluation protocol combining ranking metrics with an LLM-as-a-Judge system.
• Score: 24.718280040158245
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Forecasting future links is a central task in temporal graph (TG) reasoning, requiring models to leverage historical interactions to predict upcoming ones. Traditional neural approaches, such as temporal graph neural networks, achieve strong performance but lack explainability and cannot be applied to unseen graphs without retraining. Recent studies have begun to explore using large language models (LLMs) for graph reasoning, but most of them are constrained to static graphs or small synthetic TGs and lack the evaluation of the quality of reasoning traces generated by LLMs. In this work, we present Reasoning-Enhanced Learning for Temporal Graphs (ReaL-TG), a reinforcement learning framework that fine-tunes LLMs to perform explainable link forecasting on real-world TGs. ReaL-TG uses outcome-based reward to encourage models to self-explore reasoning strategies from graph structure and to produce explanations that directly justify their predictions. To enable evaluation on LLM-generated reasoning traces, we propose a new evaluation protocol combining ranking metrics with an LLM-as-a-Judge system that assesses both the quality of reasoning and the impact of hallucinations. Experiments with ReaL-TG-4B, obtained by fine-tuning Qwen3-4B under our framework, show that it outperforms much larger frontier LLMs, including GPT-5 mini, on ranking metrics, while producing high-quality explanations confirmed by both the LLM judge and human evaluation.

Related papers

• Are LLMs Better GNN Helpers? Rethinking Robust Graph Learning under Deficiencies with Iterative Refinement [3.2910447933011535]
  Graph Neural Networks (GNNs) are widely adopted in Web-related applications, serving as a core technique for learning from graph-structured data.<n>In real-world scenarios, such graphs exhibit deficiencies that substantially undermine GNN performance.<n>We propose Robust Graph Learning via Retrieval-Augmented Contrastive Refinement framework.
  arXiv  Detail & Related papers  (2025-10-02T11:30:51Z)
• G-reasoner: Foundation Models for Unified Reasoning over Graph-structured Knowledge [88.82814893945077]
  Large language models (LLMs) excel at complex reasoning but remain limited by static and incomplete parametric knowledge.<n>Recent graph-enhanced RAG (GraphRAG) attempts to bridge this gap by constructing tailored graphs and enabling LLMs to reason on them.<n>G-reasoner is a unified framework that integrates graph and language foundation models for reasoning over diverse graph-structured knowledge.
  arXiv  Detail & Related papers  (2025-09-29T04:38:12Z)
• GRIL: Knowledge Graph Retrieval-Integrated Learning with Large Language Models [59.72897499248909]
  We propose a novel graph retriever trained end-to-end with Large Language Models (LLMs)<n>Within the extracted subgraph, structural knowledge and semantic features are encoded via soft tokens and the verbalized graph, respectively, which are infused into the LLM together.<n>Our approach consistently achieves state-of-the-art performance, validating the strength of joint graph-LLM optimization for complex reasoning tasks.
  arXiv  Detail & Related papers  (2025-09-20T02:38:00Z)
• Graph-R1: Incentivizing the Zero-Shot Graph Learning Capability in LLMs via Explicit Reasoning [7.1931434571877375]
  Graph Neural Networks (GNNs) are limited by fixed label spaces, while Large Language Models (LLMs) lack structural inductive biases.<n>Recent advances in Large Reasoning Models (LRMs) provide a zero-shot alternative via explicit, long chain-of-thought reasoning.<n>We propose a GNN-free approach that reformulates graph tasks--node classification, link prediction, and graph classification--as textual reasoning problems solved by LRMs.
  arXiv  Detail & Related papers  (2025-08-24T14:49:02Z)
• Weak-to-Strong GraphRAG: Aligning Weak Retrievers with Large Language Models for Graph-based Retrieval Augmented Generation [29.492846663357565]
  Graph-based retrieval-augmented generation (RAG) enables large language models (LLMs) to mitigate hallucinations.<n>This paper introduces Refined Graph-based RAG (ReG) to align weak retrievers to LLMs for graph-based RAG.<n>ReG incorporates LLM feedback to get rid of spurious signals and improve the quality of the supervision.
  arXiv  Detail & Related papers  (2025-06-26T17:40:23Z)
• 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)
• Are Large Language Models Good Temporal Graph Learners? [10.964357009640105]
  We introduce Temporal Graph Talker (TGTalker), a novel temporal graph learning framework designed for Large Language Models (LLMs)<n>TGTalker utilizes the recency bias in temporal graphs to extract relevant structural information, converted to natural language for LLMs, while leveraging temporal neighbors as additional information for prediction.<n>Across five real-world networks, TGTalker performs competitively with state-of-the-art temporal graph methods.
  arXiv  Detail & Related papers  (2025-06-03T19:34:24Z)
• G1: Teaching LLMs to Reason on Graphs with Reinforcement Learning [58.73279333365234]
  Reinforcement Learning (RL) on synthetic graph-theoretic tasks can significantly scale graph reasoning abilities.<n>With RL on Erdos, G1 obtains substantial improvements in graph reasoning, where our finetuned 3B model even outperforms Qwen2.5-72B-Instruct (24x size)<n>Our findings offer an efficient, scalable path for building strong graph reasoners by finetuning LLMs with RL on graph-theoretic tasks.
  arXiv  Detail & Related papers  (2025-05-24T04:33:41Z)
• Compile Scene Graphs with Reinforcement Learning [69.36723767339001]
  Next-token prediction is the fundamental principle for training large language models (LLMs)<n>We introduce R1-SGG, a multimodal LLM (M-LLM) initially trained via supervised fine-tuning (SFT) on the scene graph dataset.<n>We design a set of graph-centric rewards, including three recall-based variants -- Hard Recall, Hard Recall+Relax, and Soft Recall.
  arXiv  Detail & Related papers  (2025-04-18T10:46:22Z)
• Integrate Temporal Graph Learning into LLM-based Temporal Knowledge Graph Model [48.15492235240126]
  Temporal Knowledge Graph Forecasting aims to predict future events based on the observed events in history.<n>Existing methods have integrated retrieved historical facts or static graph representations into Large Language Models (LLMs)<n>We propose a novel framework TGL-LLM to integrate temporal graph learning into LLM-based temporal knowledge graph model.
  arXiv  Detail & Related papers  (2025-01-21T06:12:49Z)
• Disentangled Representation Learning with Large Language Models for Text-Attributed Graphs [57.052160123387104]
  We present the Disentangled Graph-Text Learner (DGTL) model, which is able to enhance the reasoning and predicting capabilities of LLMs for TAGs. Our proposed DGTL model incorporates graph structure information through tailored disentangled graph neural network (GNN) layers. Experimental evaluations demonstrate the effectiveness of the proposed DGTL model on achieving superior or comparable performance over state-of-the-art baselines.
  arXiv  Detail & Related papers  (2023-10-27T14:00:04Z)

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.