CUP: Curriculum Learning based Prompt Tuning for Implicit Event Argument Extraction

Authors: Jiaju Lin, Qin Chen, Jie Zhou, Jian Jin, Liang He

IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on two well-known benchmark datasets show the great advantages of our proposed approach. In particular, we outperform the state-of-the-art models in both fully-supervised and low-data scenarios.
Researcher Affiliation Academia 1School of Computer Science and Technology, East China Normal University 2School of Computer Science, Fudan University
Pseudocode No The paper describes its approach conceptually and through diagrams (Figure 2, Figure 3), but does not contain a formal pseudocode block or an explicitly labeled algorithm.
Open Source Code Yes Our code is available at https://github.com/linmou/CUP
Open Datasets Yes We conduct extensive experiments on two implicit EAE datasets, namely RAMS [Ebner et al., 2020] and Wiki Events [Li et al., 2021], which have been widely used in previous studies [Li et al., 2021; Liu et al., 2021a].
Dataset Splits Yes Detailed statistics are listed in Table 1. Data Set Split Doc Event Argument RAMS Train 3194 7329 17026 Dev 399 924 2188 Test 400 871 2023 Wiki Events Train 206 3241 4542 Dev 20 345 428 Test 20 365 556
Hardware Specification Yes The experiments are conducted on a single Nividia Ge Force RTX 3090.
Software Dependencies No The paper mentions using BART and spaCy (via a footnote with URL: 'https://spacy.io/'), but does not specify version numbers for these or other software dependencies.
Experiment Setup Yes For the four stages of curriculum learning, the learning rate decreases as the learning process goes on, which is set as 1e-4, 5e-5, 3e-5 and 2e-5 respectively. The best batch size is searched from the set of {16, 32} for each stage. The hyper-parameter of α that balances the two losses in Formula (10) ranges from 0 to 1 with a step of 0.1, and we set it to 0.7 as a reliable setting. All other parameters are optimized by the Adam [Kingma and Ba, 2014] algorithm.