A Generalized Idiom Usage Recognition Model Based on Semantic Compatibility

Authors: Changsheng Liu, Rebecca Hwa6738-6745

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

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct experiments on two benchmark idiom corpora; results suggest that the proposed generalized model achieves competitive results compared to state-of-the-art per-idiom models.
Researcher Affiliation Academia Changsheng Liu, Rebecca Hwa Computer Science Department University of Pittsburgh Pittsburgh, PA 15260, USA {changsheng, hwa}@cs.pitt.edu
Pseudocode No The paper describes its model using figures and mathematical equations but does not include any pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any links or explicit statements about releasing source code for the described methodology.
Open Datasets Yes We conduct experiments on two benchmark idiom corpora: Sem Eval 2013 Task 5B corpus (Korkontzelos et al. 2013) and Verb Noun Combination (VNC) dataset (Cook, Fazly, and Stevenson 2008).
Dataset Splits No The paper mentions picking '10 idioms that are different from the evaluation set' to tune a bias term ('bu'), implying a development set, but it does not provide specific details on standard training/validation/test splits (e.g., percentages or sample counts) for the main experimental setup. It mentions '5-fold cross validation' for baselines, but not for its own model's splits.
Hardware Specification No The paper describes the training process and hyperparameters but does not provide specific details about the hardware used for running the experiments (e.g., CPU/GPU models, memory, or cloud instances).
Software Dependencies No The paper mentions using the 'Adam optimizer' and 'word2vec' but does not specify version numbers for any software dependencies, libraries, or programming languages.
Experiment Setup Yes The hyperparameters are summarized in Table 1. Parameter Value word embedding size 200 context embedding size 200 LSTM hidden size 200 f1 input/output size 200/400 f2 input/output size 400/200 negative samples 15 epoch 10 batch size 500 learning rate 0.001