Graph-based Neural Sentence Ordering
Authors: Yongjing Yin, Linfeng Song, Jinsong Su, Jiali Zeng, Chulun Zhou, Jiebo Luo
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results show that our proposed model outperforms the existing stateof-the-art systems on several benchmark datasets, demonstrating the effectiveness of our model. |
| Researcher Affiliation | Collaboration | 1Xiamen University, Xiamen, China 2University of Rochester, Rochester, NY, U.S. 3Tencent AI Lab, Bellevue, WA, U.S. |
| Pseudocode | No | No structured pseudocode or algorithm blocks are present in the paper. The model is described through mathematical equations and text. |
| Open Source Code | Yes | Our code is available at https://github.com/DeepLearnXMU/NSEG.git. |
| Open Datasets | Yes | SIND. It has 50K stories for the visual storytelling task1, which is in a different domain from the others. Here we use each story as a paragraph. 1http://visionandlanguage.net/VIST/ |
| Dataset Splits | No | The paper mentions using a 'validation set' (e.g., 'arXive Abstract validation set') and 'test set' in its results, but does not provide specific split percentages, sample counts, or explicit methodology for partitioning the data into training, validation, and test sets. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running the experiments. |
| Software Dependencies | No | The paper mentions using NLTK and Stanford Parser, but does not provide specific version numbers for these or any other software libraries or dependencies used in the experiments. |
| Experiment Setup | Yes | Settings. Our settings follow Cui et al., [2018] for fair comparison. We use 100-dimension Glove word embeddings3. The hidden size of LSTM is 300 for NIPS Abstract and 512 for the others. For our GRN encoders, The state sizes for sentence and entity nodes are set to 512 and 150, respectively. The size of edge embeddings is set to 50. Adadelta [Zeiler, 2012] is adopted as the optimizer with ϵ = 10 6, ρ = 0.95 and initial learning rate 1.0. For regularization term, we employ L2 weight decay with coefficient 10 5 and dropout with probability 0.5. Batch size is 16 for training and beam search with size 64 is implemented for decoding. |