# GCoT: Chain-of-Thought Prompt Learning for Graphs

arXiv: https://arxiv.org/abs/2502.08092 · HTML: https://mlsanity.com/paper/2502.08092/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 43 of its 60 references that are in the map.
- First posted on arXiv 12 Feb 2025.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 30 of 43)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [How Powerful are Graph Neural Networks?](https://mlsanity.com/paper/1810.00826/) (2018)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [A Comprehensive Survey on Graph Neural Networks](https://mlsanity.com/paper/1901.00596/) (2019)
- [Graph Neural Networks: A Review of Methods and Applications](https://mlsanity.com/paper/1812.08434/) (2018)
- [Conditional Prompt Learning for Vision-Language Models](https://mlsanity.com/paper/2203.05557/) (2022)
- [Variational Graph Auto-Encoders](https://mlsanity.com/paper/1611.07308/) (2016)
- [Graph Contrastive Learning with Augmentations](https://mlsanity.com/paper/2010.13902/) (2020)
- [Deep Graph Infomax](https://mlsanity.com/paper/1809.10341/) (2018)
- [Pitfalls of Graph Neural Network Evaluation](https://mlsanity.com/paper/1811.05868/) (2018)
- [Strategies for Pre-training Graph Neural Networks](https://mlsanity.com/paper/1905.12265/) (2019)
- [Geom-GCN: Geometric Graph Convolutional Networks](https://mlsanity.com/paper/2002.05287/) (2020)
- [GPT Understands, Too](https://mlsanity.com/paper/2103.10385/) (2021)
- [Benchmarking Graph Neural Networks](https://mlsanity.com/paper/2003.00982/) (2020)
- [PAL: Program-aided Language Models](https://mlsanity.com/paper/2211.10435/) (2022)
- [Multi-scale Attributed Node Embedding](https://mlsanity.com/paper/1909.13021/) (2019)
- [Graph Transformer Networks](https://mlsanity.com/paper/1911.06455/) (2019)
- [GCC: Graph Contrastive Coding for Graph Neural Network Pre-Training](https://mlsanity.com/paper/2006.09963/) (2020)
- [GPT-GNN: Generative Pre-Training of Graph Neural Networks](https://mlsanity.com/paper/2006.15437/) (2020)
- [Towards Revealing the Mystery behind Chain of Thought: A Theoretical Perspective](https://mlsanity.com/paper/2305.15408/) (2023)
- [GraphPrompt: Unifying Pre-Training and Downstream Tasks for Graph Neural Networks](https://mlsanity.com/paper/2302.08043/) (2023)
- [All in One: Multi-task Prompting for Graph Neural Networks](https://mlsanity.com/paper/2307.01504/) (2023)
- [Graph Chain-of-Thought: Augmenting Large Language Models by Reasoning on Graphs](https://mlsanity.com/paper/2404.07103/) (2024)
- [Augmenting Low-Resource Text Classification with Graph-Grounded Pre-training and Prompting](https://mlsanity.com/paper/2305.03324/) (2023)
- [SAMGPT: Text-free Graph Foundation Model for Multi-domain Pre-training and Cross-domain Adaptation](https://mlsanity.com/paper/2502.05424/) (2025)
- [HGPROMPT: Bridging Homogeneous and Heterogeneous Graphs for Few-shot Prompt Learning](https://mlsanity.com/paper/2312.01878/) (2023)
- [Text-Free Multi-domain Graph Pre-training: Toward Graph Foundation Models](https://mlsanity.com/paper/2405.13934/) (2024)

## Built on by (most cited first, 9 of 9)

- [PRISM-Bench: A Benchmark of Puzzle-Based Visual Tasks with CoT Error Detection](https://mlsanity.com/paper/2510.23594/) (2025)
- [Beyond Single-Granularity Prompts: A Multi-Scale Chain-of-Thought Prompt Learning for Graph](https://mlsanity.com/paper/2510.09394/) (2025)
- [Dynamic Graph Prompting via Topology-Routed Mixed-Curvature Experts](https://mlsanity.com/paper/2608.06031/) (2026)
- [Autonomous Question Formation for Large Language Model-Driven AI Systems](https://mlsanity.com/paper/2602.01556/) (2026)
- [GraphReAct: Reasoning and Acting for Multi-step Graph Inference](https://mlsanity.com/paper/2605.07357/) (2026)
- [MolGA: Molecular Graph Adaptation with Pre-trained 2D Graph Encoder](https://mlsanity.com/paper/2510.07289/) (2025)
- [GRASP: Graph Reasoning Agents for Systems Pharmacology with Human-in-the-Loop](https://mlsanity.com/paper/2512.05502/) (2025)
- [GraphVec: Cross-Domain Graph Vectorization for Graph-Level Representation Learning](https://mlsanity.com/paper/2602.04244/) (2026)
- [Decompose and Transfer: CoT-Prompting Enhanced Alignment for Open-Vocabulary Temporal Action Detection](https://mlsanity.com/paper/2603.24030/) (2026)

## Abstract (arXiv)

Chain-of-thought (CoT) prompting has achieved remarkable success in natural language processing (NLP). However, its vast potential remains largely unexplored for graphs. This raises an interesting question: How can we design CoT prompting for graphs to guide graph models to learn step by step? On one hand, unlike natural languages, graphs are non-linear and characterized by complex topological structures. On the other hand, many graphs lack textual data, making it difficult to formulate language-based CoT prompting. In this work, we propose the first CoT prompt learning framework for text-free graphs, GCoT. Specifically, we decompose the adaptation process for each downstream task into a series of inference steps, with each step consisting of prompt-based inference, ``thought'' generation, and thought-conditioned prompt learning. While the steps mimic CoT prompting in NLP, the exact mechanism differs significantly. Specifically, at each step, an input graph, along with a prompt, is first fed into a pre-trained graph encoder for prompt-based inference. We then aggregate the hidden layers of the encoder to construct a ``thought'', which captures the working state of each node in the current step. Conditioned on this thought, we learn a prompt specific to each node based on the current state. These prompts are fed into the next inference step, repeating the cycle. To evaluate and analyze the effectiveness of GCoT, we conduct comprehensive experiments on eight public datasets, which demonstrate the advantage of our approach.