# Global-Recent Semantic Reasoning on Dynamic Text-Attributed Graphs with Large Language Models

arXiv: https://arxiv.org/abs/2509.18742 · HTML: https://mlsanity.com/paper/2509.18742/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 18 of its 62 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)

## Builds on (most cited first, 18 of 18)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Phi-4 Technical Report](https://mlsanity.com/paper/2412.08905/) (2024)
- [Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networks](https://mlsanity.com/paper/2007.02901/) (2020)
- [Harnessing Explanations: LLM-to-LM Interpreter for Enhanced Text-Attributed Graph Representation Learning](https://mlsanity.com/paper/2305.19523/) (2023)
- [SimTeG: A Frustratingly Simple Approach Improves Textual Graph Learning](https://mlsanity.com/paper/2308.02565/) (2023)
- [TCL: Transformer-based Dynamic Graph Modelling via Contrastive Learning](https://mlsanity.com/paper/2105.07944/) (2021)
- [GraphEdit: Large Language Models for Graph Structure Learning](https://mlsanity.com/paper/2402.15183/) (2024)
- [Exploring the Potential of Large Language Models for Heterophilic Graphs](https://mlsanity.com/paper/2408.14134/) (2024)
- [LLM-driven Knowledge Distillation for Dynamic Text-Attributed Graphs](https://mlsanity.com/paper/2502.10914/) (2025)
- [Unifying Text Semantics and Graph Structures for Temporal Text-attributed Graphs with Large Language Models](https://mlsanity.com/paper/2503.14411/) (2025)
- [Efficient End-to-end Language Model Fine-tuning on Graphs](https://mlsanity.com/paper/2312.04737/) (2023)
- [GraphiT: Efficient Node Classification on Text-Attributed Graphs with Prompt Optimized LLMs](https://mlsanity.com/paper/2502.10522/) (2025)
- [Exploring the Potential of Large Language Models as Predictors in Dynamic Text-Attributed Graphs](https://mlsanity.com/paper/2503.03258/) (2025)
- [Integrating Structural and Semantic Signals in Text-Attributed Graphs with BiGTex](https://mlsanity.com/paper/2504.12474/) (2025)

## Built on by (most cited first, 2 of 2)

- [Self-Evolving Agents as Dynamic Graph Transformation: A Survey and New Perspective](https://mlsanity.com/paper/2608.18104/) (2026)
- [Transfer Learning for Edge Classification on Dynamic Text-Attributed Graphs](https://mlsanity.com/paper/2609.32849/) (2026)

## Abstract (arXiv)

Dynamic Text-Attribute Graphs (DyTAGs), characterized by time-evolving graph interactions and associated text attributes, are prevalent in real-world applications. Existing methods, such as Graph Neural Networks (GNNs) and Large Language Models (LLMs), mostly focus on static TAGs. Extending these existing methods to DyTAGs is challenging as they largely neglect the recent-global temporal semantics: the recent semantic dependencies among interaction texts and the global semantic evolution of nodes over time. Furthermore, applying LLMs to the abundant and evolving text in DyTAGs faces efficiency issues. To tackle these challenges, we propose Dynamic Global-Recent Adaptive Semantic Processing (DyGRASP), a novel method that leverages LLMs and temporal GNNs to efficiently and effectively reason on DyTAGs. Specifically, we first design a node-centric implicit reasoning method together with a sliding window mechanism to efficiently capture recent temporal semantics. In addition, to capture global semantic dynamics of nodes, we leverage explicit reasoning with tailored prompts and an RNN-like chain structure to infer long-term semantics. Lastly, we intricately integrate the recent and global temporal semantics as well as the dynamic graph structural information using updating and merging layers. Extensive experiments on DyTAG benchmarks demonstrate DyGRASP's superiority, achieving up to 34% improvement in Hit@10 for destination node retrieval task. Besides, DyGRASP exhibits strong generalization across different temporal GNNs and LLMs.