Adapting BERT for Target-Oriented Multimodal Sentiment Classification

Authors: Jianfei Yu, Jing Jiang

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that our model can outperform several highly competitive approaches for TSC and TMSC.
Researcher Affiliation Academia 1School of Computer Science and Engineering, Nanjing University of Science and Technology, China 2School of Information Systems, Singapore Management University, Singapore {jfyu, jingjiang}@smu.edu.sg
Pseudocode No The paper describes the methodology using textual explanations and mathematical equations, but no structured pseudocode or algorithm blocks are provided.
Open Source Code No We make our annotations for the two TMSC datasets publicly available via the link: https://github.com/jeffery Yu/Tom BERT. (This link is specifically for annotations, not the model's source code, making it insufficient for 'open_source_code').
Open Datasets Yes We make our annotations for the two TMSC datasets publicly available via the link: https://github.com/jeffery Yu/Tom BERT. ... LAPTOP and REST from Sem Eval-2014 Task 4 [Pontiki et al., 2014] as well as TWITTER-14 constructed by [Dong et al., 2014]
Dataset Splits Yes TWITTER-15 TWITTER-17 ... Dev. 303 149 670 1122 ... 515 144 517 1176 ... We build our Tom BERT model on top of the pre-trained uncased BERTbase model released by [Devlin et al., 2018], and tune the hyper-parameters on the development set of each dataset.
Hardware Specification Yes All the models are fine-tuned for 8 epochs, and are implemented based on Py Torch with a NVIDIA Tesla V100 GPU.
Software Dependencies No The models are implemented based on Py Torch, but no specific version number for Py Torch or other software dependencies is provided.
Experiment Setup Yes Specifically, for BERT-based models, we set the learning rate as 5e-5, the number of attention heads as m = 12, and the dropout rate as 0.1. The batch size is respectively set as 16 and 32 for all the models for TSC and TMSC, respectively. Besides, for Tom BERT, the maximum length of the sentence input and the target input are respectively set as N = 64 and M = 16. The number of layers for encoding the sentence input and the target input are both set to be 12, i.e., Ls = Lc = 12, where the parameters are both initialized from the pre-trained BERTbase model. All the models are fine-tuned for 8 epochs