Abstract:Heterogeneous graph learning aims to capture complex relationships and diverse relational semantics among entities in a heterogeneous graph to obtain meaningful representations for nodes and edges. Recent advancements in heterogeneous graph neural networks (HGNNs) have achieved state-of-the-art performance by considering relation heterogeneity and using specialized message functions and aggregation rules. However, existing frameworks for heterogeneous graph learning have limitations in generalizing across diverse heterogeneous graph datasets. Most of these frameworks follow the "pre-train" and "fine-tune" paradigm on the same dataset, which restricts their capacity to adapt to new and unseen data. This raises the question: "Can we generalize heterogeneous graph models to be well-adapted to diverse downstream learning tasks with distribution shifts in both node token sets and relation type heterogeneity?'' To tackle those challenges, we propose HiGPT, a general large graph model with Heterogeneous graph instruction-tuning paradigm. Our framework enables learning from arbitrary heterogeneous graphs without the need for any fine-tuning process from downstream datasets. To handle distribution shifts in heterogeneity, we introduce an in-context heterogeneous graph tokenizer that captures semantic relationships in different heterogeneous graphs, facilitating model adaptation. We incorporate a large corpus of heterogeneity-aware graph instructions into our HiGPT, enabling the model to effectively comprehend complex relation heterogeneity and distinguish between various types of graph tokens. Furthermore, we introduce the Mixture-of-Thought (MoT) instruction augmentation paradigm to mitigate data scarcity by generating diverse and informative instructions. Through comprehensive evaluations, our proposed framework demonstrates exceptional performance in terms of generalization performance.

HeGTa: Leveraging Heterogeneous Graph-enhanced Large Language Models for Few-shot Complex Table Understanding

HGT: Leveraging Heterogeneous Graph-enhanced Large Language Models for Few-shot Complex Table Understanding

TableGPT: Towards Unifying Tables, Nature Language and Commands into One GPT

TableGPT: Few-shot Table-to-Text Generation with Table Structure Reconstruction and Content Matching

TabPedia: Towards Comprehensive Visual Table Understanding with Concept Synergy

Tree-of-Table: Unleashing the Power of LLMs for Enhanced Large-Scale Table Understanding

TabPrompt: Graph-based Pre-training and Prompting for Few-shot Table Understanding.

Table Meets LLM: Can Large Language Models Understand Structured Table Data? A Benchmark and Empirical Study

MTU-Bench: A Multi-granularity Tool-Use Benchmark for Large Language Models

TEGEE: Task dEfinition Guided Expert Ensembling for Generalizable and Few-shot Learning

HiGPT: Heterogeneous Graph Language Model

Multimodal Table Understanding

GraphOTTER: Evolving LLM-based Graph Reasoning for Complex Table Question Answering

TabLLM: Few-shot Classification of Tabular Data with Large Language Models

Table-GPT: Table-tuned GPT for Diverse Table Tasks

Bridge the Gap between Language models and Tabular Understanding

GraphGPT: Graph Instruction Tuning for Large Language Models

Enhancing Table Recognition with Vision LLMs: A Benchmark and Neighbor-Guided Toolchain Reasoner

Pretraining Language Models with Text-Attributed Heterogeneous Graphs

TableRobot: an Automatic Annotation Method for Heterogeneous Tables