Hierarchical Macro Discourse Parsing Based on Topic Segmentation
Authors: Feng Jiang, Yaxin Fan, Xiaomin Chu, Peifeng Li, Qiaoming Zhu, Fang Kong13152-13160
AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The experimental results on both Chinese MCDTB and English RST-DT show that our proposed method outperforms the state-of-the-art baselines significantly. |
| Researcher Affiliation | Academia | Feng Jiang, Yaxin Fan, Xiaomin Chu, Peifeng Li, Qiaoming Zhu , Fang Kong School of Computer Science and Technology, Soochow University, China {fjiang, yxfansupery}@stu.suda.edu.cn {xmchu, pfli, qmzhu, kongfang}@suda.edu.cn |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper refers to third-party open-source code and libraries used ('https://github.com/yizhongw/Stage DP', 'https://github.com/Cyber ZHG/keras-bert') but does not state that their own implementation code for MDParser-TS is open-source or provide a link to it. |
| Open Datasets | Yes | In discourse parsing, our experiments are evaluated on the Macro Chinese Discourse Treebank (MCDTB) (Jiang et al. 2018b) that contains 720 news annotated document-level macro discourse trees. English RST-DT (Carlson, Marcu, and Okurowski 2003) is one of the popular discourse corpora (Subba and Di Eugenio 2009; Zeldes 2017) that annotates the discourse structure, nuclearity, and relationship for the whole document. |
| Dataset Splits | Yes | There are 80% data for the training set and 20% data for the test set. In particular, to balance the training set and the test set, we divide the documents containing different numbers of paragraphs into the training set (576 documents) and the test set (144 documents) according to the proportion. Finally, there are 3194 paragraphs in the training set and 791 paragraphs in the test set, and we randomly select 10% of the training set as the validation set. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper mentions using 'Keras library' and 'keras bert package' but does not specify their version numbers or other software dependencies with specific versions. |
| Experiment Setup | Yes | The key hyper parameters are following: batch-size=2, epoch=5, max-length=512, and learning-rate=1e-5. |