Learning Modality-Specific Representations with Self-Supervised Multi-Task Learning for Multimodal Sentiment Analysis

Authors: Wenmeng Yu, Hua Xu, Ziqi Yuan, Jiele Wu10790-10797

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Last, we conduct extensive experiments on three public multimodal baseline datasets. The experimental results validate the reliability and stability of auto-generated unimodal supervisions. On MOSI and MOSEI datasets, our method surpasses the current state-of-the-art methods.
Researcher Affiliation Academia Wenmeng Yu, Hua Xu, Ziqi Yuan, Jiele Wu State Key Laboratory of Intelligent Technology and Systems, Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China ywm18@mails.tsinghua.edu.cn, xuhua@mail.tsinghua.edu.cn, ziqiyuan@bupt.edu.cn, 1120171196@bit.edu.cn
Pseudocode Yes Algorithm 1 Unimodal Supervisions Update Policy
Open Source Code Yes The full codes are available at https://github.com/thuiar/Self-MM.
Open Datasets Yes In this work, we use three public multimodal sentiment analysis datasets, MOSI (Zadeh et al. 2016), MOSEI (Zadeh et al. 2018b), and SIMS (Yu et al. 2020a).
Dataset Splits Yes Table 2: Dataset statistics in MOSI, MOSEI, and SIMS. # Train # Valid # Test # All
Hardware Specification No The paper mentions experimental settings like optimizer and learning rates, but it does not specify any hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions using a 'pre-trained 12-layers BERT' and 's LSTM' but does not provide specific version numbers for any software dependencies or libraries like PyTorch, TensorFlow, etc.
Experiment Setup Yes We use Adam as the optimizer and use the initial learning rate of 5e 5 for Bert and 1e 3 for other parameters.