README.md

metadata

license: apache-2.0

Overview

This dataset covers the encoder embeddings and prediction results of LLMs of paper 'Model Generalization on Text Attribute Graphs: Principles with Lagre Language Models', Haoyu Wang, Shikun Liu, Rongzhe Wei, Pan Li.

Dataset Description

The dataset structure should be organized as follows:

/dataset/
│── [dataset_name]/
│   │── processed_data.pt    # Contains labels and graph information
│   │── [encoder]_x.pt       # Features extracted by different encoders
│   │── categories.csv       # label name raw texts
│   │── raw_texts.pt       # raw text of each node

File Descriptions

• processed_data.pt: A PyTorch file storing the processed dataset, including graph structure and node labels. Note that in heterophilic datasets, thie is named as [Dataset].pt, where Dataset could be Cornell, etc, and should be opened with DGL.
• [encoder]_x.pt: Feature matrices extracted using different encoders, where [encoder] represents the encoder name.
• categories.csv: raw label names.
• raw_texts.pt: raw node texts. Note that in heterophilic datasets, this is named as [Dataset].csv, where Dataset can be Cornell, etc.

Dataset Naming Convention

[dataset_name] should be one of the following:

• cora
• citeseer
• pubmed
• bookhis
• bookchild
• sportsfit
• wikics
• cornell
• texas
• wisconsin
• washington

Encoder Naming Convention

[encoder] can be one of the following:

• sbert (the sentence-bert encoder)
• roberta (the Roberta encoder)
• llmicl_primary (the vanilla LLM2Vec)
• llmicl_class_aware (the task-adaptive encoder)
• llmgpt_text-embedding-3-large (the embedding api text-embedding-3-large by openai)

Results Description

The ./results/ folder consists of prediction results of GPT-4o in node text classification and GPT-4o-mini in homophily ratio prediction.

./results/nc_[DATASET]/4o/llm_baseline       # node text prediction
./results/nc_[DATASET]/4o_mini/agenth        # homophily ratio prediction