Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
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.
TEILP: Time Prediction over Knowledge Graphs via Logical Reasoning
Authors: Siheng Xiong, Yuan Yang, Ali Payani, James C Kerce, Faramarz Fekri
AAAI 2024 | Venue PDF | LLM Run Details | Input Tokens: 17,251 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 2,454 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We compare TEILP with state-of-the-art methods on five benchmark datasets. We show that our model achieves a significant improvement over baselines while providing interpretable explanations. In particular, we consider several scenarios where training samples are limited, event types are imbalanced, and forecasting the time of future events based on only past events is desired. In all these cases, TEILP outperforms state-of-the-art methods in terms of robustness. ... Experiments Datasets We evaluate the proposed method TEILP on five benchmark temporal knowledge graph datasets: WIKIDATA12k, YAGO11k (Dasgupta, Ray, and Talukdar 2018), ICEWS14, ICEWS05-15 (Garc ıa-Dur an, Dumanˇci c, and Niepert 2018), and GDELT100 (Leetaru and Schrodt 2013). ... Evaluation Metrics For interval-based datasets, we adopt a new evaluation metric ae IOU, proposed by (Jain et al. 2020). ... The results of the experiments are shown in Table 1, where TEILP outperforms all baselines with respect to all metrics. |
| Researcher Affiliation | Collaboration | 1 Georgia Institute of Technology, 225 North Avenue, Atlanta, Georgia 30332 USA 2 Cisco Systems Inc, 300 East Tasman Dr, San Jose, California 95134 USA EMAIL, EMAIL, EMAIL, EMAIL |
| Pseudocode | Yes | Algorithm 1 : Rule Learning ... Algorithm 2 : Rule Application |
| Open Source Code | Yes | 1Code and data available at https://github.com/xiongsiheng/ TEILP. |
| Open Datasets | Yes | We evaluate the proposed method TEILP on five benchmark temporal knowledge graph datasets: WIKIDATA12k, YAGO11k (Dasgupta, Ray, and Talukdar 2018), ICEWS14, ICEWS05-15 (Garc ıa-Dur an, Dumanˇci c, and Niepert 2018), and GDELT100 (Leetaru and Schrodt 2013). ... In the supplementary material, we provide a detailed introduction and dataset statistics. |
| Dataset Splits | Yes | To ensure a fair comparison, we use the split provided by (Jain et al. 2020) for WIKIDATA12k, YAGO11k, ICEWS14, ICEWS05-15 datasets and (Goel et al. 2020) for GDELT dataset. ... In the low-data scenario, given the same validation and test set, we change the size of the training set over a broad range. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU, GPU models, or memory) used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers. |
| Experiment Setup | No | The paper mentions some settings like 'The maximum rule length of our method is set to 5 for YAGO11k, and 3 for the others' and that 'During training, functions f( ) will be fitted, and weights a and w will be learned' and 'Training of the model is to minimize the log-likelihood loss', but it does not provide detailed hyperparameters (e.g., learning rate, batch size, optimizer details) or a dedicated 'Experimental Setup' section. |