Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates.
Multi-Document Transformer for Personality Detection
Authors: Feifan Yang, Xiaojun Quan, Yunyi Yang, Jianxing Yu14221-14229
AAAI 2021 | Venue PDF | LLM Run Details | Input Tokens: 17,164 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 3,464 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the proposed model on the Kaggle and Pandora MBTI datasets and the experimental results show that it compares favorably with baseline methods. |
| Researcher Affiliation | Academia | Feifan Yang, Xiaojun Quan*, Yunyi Yang, Jianxing Yu School of Data and Computer Science, Sun Yat-sen University, China EMAIL, EMAIL |
| Pseudocode | No | The paper includes architectural diagrams (Figure 2 and Figure 3) and mathematical equations, but no structured pseudocode or algorithm blocks. |
| Open Source Code | No | No statement or link indicating that the source code for the methodology is openly available. |
| Open Datasets | Yes | Following previous studies (Hernandez and Knight 2017; Keh, Cheng et al. 2019; Gjurkovi c et al. 2020), we conduct experiments on the Kaggle2 and Pandora3 MBTI personality datasets. ... 2https://www.kaggle.com/datasnaek/mbti-type 3https://psy.takelab.fer.hr/datasets/all/ |
| Dataset Splits | Yes | Then, we randomly split them into a 60-20-20 proportion for training, validation, and testing, respectively. |
| Hardware Specification | Yes | We use Pytorch (Paszke et al. 2019) to implement all the deep learning models on four 2080Ti GPU cards. |
| Software Dependencies | No | The paper mentions "Pytorch (Paszke et al. 2019)", "Adam (Kingma and Ba 2014) optimizer", "bert-base-cased (Devlin et al. 2018)" and "xlnet-base-cased (Yang et al. 2019b)", but only PyTorch has a parenthetical reference that *could* imply a version, but it's a citation, not an explicit version number. No other software versions are provided for reproducibility. |
| Experiment Setup | Yes | For training, we use the Adam (Kingma and Ba 2014) optimizer with an initial learning rate α = 2e-5 and a mini-batch size of 24. Following previous work, we set the max number of posts to 50 for each user and the max length to 70 for each post. ... The hidden size dt of our dimension attention module is set to 768. |