Abstract Rule Learning for Paraphrase Generation

Authors: Xianggen Liu, Wenqiang Lei, Jiancheng Lv, Jizhe Zhou

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experimental results demonstrate the superiority of RULER over previous state-of-the-art methods in terms of paraphrase quality, generalization ability and interpretability. We evaluate the effectiveness of our method on two benchmark paraphrasing datasets, namely, the Quora question pairs and Wikianswers datasets.
Researcher Affiliation Academia College of Computer Science, Sichuan University {liuxianggen,lvjiancheng,jzzhou}@scu.edu.cn, wenqianglei@gmail.com
Pseudocode No The paper does not contain any clearly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statement about open-sourcing the code or a link to a code repository.
Open Datasets Yes We evaluate RULER on two widely used datasets, namely, the Quora question pairs and Wikianswers datasets. The Wikianswers dataset [Fader et al., 2013] comprises 2.3M pairs of question paraphrases scraped from the Wikianswers website.
Dataset Splits Yes On these two datasets, we adopt the same data splits with Hosking and Lapata [2021] for a fair comparison.
Hardware Specification No The paper does not specify any hardware details (e.g., CPU, GPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions software like "Transformer architecture" and "SEPARATOR", but does not provide specific version numbers for these or other software dependencies.
Experiment Setup Yes The paraphrase generators PGα and PGβ adopt the Transformer architecture and the previous best performing paraphraser model (i.e., SEPARATOR), respectively. To have a fair comparison with SEPARATOR, we use the same hyperparameters with it. ... The minimum number Cmin of matched samples was set to 16. The improvement threshold τ of the generator loss is 0.2.