Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
TRACI: A Data-centric Approach for Multi-Domain Generalization on Graphs
Authors: Yusheng Zhao, Changhu Wang, Xiao Luo, Junyu Luo, Wei Ju, Zhiping Xiao, Ming Zhang
AAAI 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on various benchmark datasets demonstrate that TRACI outperforms state-of-the-art baselines, validating its effectiveness. |
| Researcher Affiliation | Academia | Yusheng Zhao1, Changhu Wang2, Xiao Luo3,*, Junyu Luo1, Wei Ju1, Zhiping Xiao4, *, Ming Zhang1* 1State Key Laboratory for Multimedia Information Processing, School of Computer Science, PKU-Anker LLM Lab, Peking University, Beijing, China 2Department of Statistics, University of California, Los Angeles, CA, USA 3Department of Computer Science, University of California, Los Angeles, CA, USA 4Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA EMAIL, EMAIL, EMAIL, EMAIL, EMAIL |
| Pseudocode | No | The paper describes methods using text and mathematical equations, but it does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | For details, see https://github.com/YushengZhao/TRACI-AAAI25/. |
| Open Datasets | Yes | Two benchmarks are used in the experiments: Citations and Proteins. The Citations benchmark consists of three real-world networks from (Tang et al. 2008), i.e. ACMv9, Citationv1 and DBLPv7, collected by different sources during different periods. The Proteins benchmark consists of four protein datasets from the PDB database (Berman et al. 2000) with their single-chain data as graphs. |
| Dataset Splits | Yes | We employ the leave-one-domain-out protocol (Dubey et al. 2021). |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers. |
| Experiment Setup | Yes | For hyperparameters, is set to 0.6, ! is 0.02, λ in Eq.14 is 0.1, and L is 2. |