GSN: A Graph-Structured Network for Multi-Party Dialogues
Authors: Wenpeng Hu, Zhangming Chan, Bing Liu, Dongyan Zhao, Jinwen Ma, Rui Yan
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 4 Experiments, 4.1 Experimental Setups, 4.3 Results and Analysis, Experimental results show that GSN significantly outperforms existing sequence-based models. |
| Researcher Affiliation | Academia | 1Department of Information Science, School of Mathematical Sciences, Peking University 2Center for Data Science, Peking University 3Institute of Computer Science and Technology, Peking University 4Department of Computer Science, University of Illinois at Chicago |
| Pseudocode | Yes | we reformulate it as matrix operations (also see UG-E in Figure 2) and give the pseudo-code in Algorithm 1. |
| Open Source Code | Yes | The code of our model can be found here 4. 4https://github.com/morning-dews/GSN-Dialogues |
| Open Datasets | Yes | Our experiment uses the Ubuntu Dialogue Corpus2 [Lowe et al., 2015] as it is the only benchmark corpus with annotated multiple interlocutors. 2http://dataset.cs.mcgill.ca/ubuntu-corpus-1.0/ |
| Dataset Splits | Yes | We randomly divide the corpus into the training, development (with 5k q/a pairs), and test (with 5k q/a pairs) sets. |
| Hardware Specification | Yes | We run all experiments on a single GTX Titan X GPU, and training takes 25 epochs. |
| Software Dependencies | No | The paper mentions software components like GRU units and the Adam algorithm but does not provide specific version numbers for these or other software dependencies. |
| Experiment Setup | Yes | The number of hidden units is set as 300 and the word embedding dimension is set as 300. We have 2 layers for both word-level encoder and decoder. The network parameters are updated using the Adam algorithm [Kingma and Ba, 2014] with the learning rate of 0.0001. All utterances are clipped to 30 words. We run all experiments on a single GTX Titan X GPU, and training takes 25 epochs. We set α in Eq. 4 to 0.25. |