Hierarchical Aligned Multimodal Learning for NER on Tweet Posts

Authors: Peipei Liu, Hong Li, Yimo Ren, Jie Liu, Shuaizong Si, Hongsong Zhu, Limin Sun

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct experiments on two open datasets, and the results and detailed analysis demonstrate the advantage of our model.
Researcher Affiliation Academia 1Institute of Information Engineering, Chinese Academy of Sciences 19 Shucun Road, Haidian District, Beijing 100085 P.R.China 2School of Cyber Security, University of Chinese Academy of Sciences 19 Yuquan Road, Shijingshan District, Beijing 100049 P.R.China
Pseudocode No The paper includes mathematical equations and architectural diagrams (e.g., Figure 2) but does not provide structured pseudocode or algorithm blocks.
Open Source Code No The paper does not include an unambiguous statement or a direct link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes The experiments are carried out on the datasets TWITTER2015 and TWITTER-2017, which are constructed based on Twitter by (Lu et al. 2018) and (Zhang et al. 2018) separately.
Dataset Splits No The paper states, 'The experiments are carried out on the datasets TWITTER2015 and TWITTER-2017' and mentions 'Implementation Details' and 'Appendices' for hyperparameters, but it does not explicitly specify the training/validation/test dataset splits (e.g., percentages, counts, or a referenced split methodology) in the provided text.
Hardware Specification No The paper does not provide any specific hardware details such as GPU models, CPU types, or memory specifications used for running the experiments.
Software Dependencies No The paper mentions using models like BERT, ResNet, Faster-RCNN, and general components like Transformer and CRF layer, but it does not provide specific version numbers for any software libraries or dependencies used in the implementation.
Experiment Setup No The paper states, 'For both datasets, we have the same hyperparameters and the specific parameter content can be found in the Appendices.' This indicates that the detailed experimental setup and hyperparameter values are deferred to an external appendix, not explicitly provided within the main body of the paper.