HGPrompt: Bridging Homogeneous and Heterogeneous Graphs for Few-Shot Prompt Learning

Authors: Xingtong Yu, Yuan Fang, Zemin Liu, Xinming Zhang

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we thoroughly evaluate and analyze HGPROMPT through extensive experiments on three public datasets.
Researcher Affiliation Academia 1University of Science and Technology of China, China 2 Singapore Management University, Singapore 3 National University of Singapore, Singapore
Pseudocode No No explicit pseudocode or algorithm blocks are provided in the main text, only mathematical formulas and descriptions of the framework components.
Open Source Code No The paper does not provide any explicit statement about open-sourcing the code for the described methodology, nor does it include a link to a code repository.
Open Datasets Yes We conduct experiments on three benchmark datasets. (1) ACM serves as a citation network... (2) DBLP serves as an all-encompassing bibliographic database... (3) Freebase (Bollacker et al. 2008)... For all these datasets, we employ the same raw data as Simple-HGN (Lv et al. 2021). We provide a summary of these datasets in Table 1.
Dataset Splits Yes we randomly generate 100 one-shot tasks for model training and validation.
Hardware Specification No The paper does not explicitly describe the specific hardware (CPU, GPU models, memory, etc.) used for running the experiments.
Software Dependencies No We adopt Micro F and Macro F (Pedregosa et al. 2011; Lv et al. 2021) as the evaluation metrics.
Experiment Setup Yes For hyperparameter settings and other implementation details about the baselines and HGPROMPT, see Appendix C.