Factual and Informative Review Generation for Explainable Recommendation

Authors: Zhouhang Xie, Sameer Singh, Julian McAuley, Bodhisattwa Prasad Majumder

AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on Yelp, Trip Advisor, and Amazon Movie Reviews dataset show our model could generate explanations that more reliably entail existing reviews, are more diverse, and are rated more informative by human evaluators.
Researcher Affiliation Academia 1 University of California, San Diego 2 University of California, Irvine {zhx022, jmcauley, bmajumde}@ucsd.edu, sameer@uci.edu
Pseudocode No The paper includes architectural diagrams (Figure 1, Figure 2) but no explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes 1https://github.com/zhouhanxie/PRAG
Open Datasets Yes We conduct our experiments on three publicly available dataset and splits from various domains (Li, Zhang, and Chen 2020): Yelp4 (restaurant reviews), Trip Advisor5 (hotel) and Amazon Movies and TV (He and Mc Auley 2016). (...) 4https://www.yelp.com/dataset/challenge 5https://www.tripadvisor.com
Dataset Splits Yes We conduct our experiments on three publicly available dataset and splits from various domains (Li, Zhang, and Chen 2020): Yelp4 (restaurant reviews), Trip Advisor5 (hotel) and Amazon Movies and TV (He and Mc Auley 2016). Note that the data splits guarantee that products in the test set always appear in the training set.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory specifications) used for running the experiments.
Software Dependencies No The paper mentions software components like MPNET, Huggingface Transformers, GPT-2, Unified QA, and T5 but does not specify their version numbers, which is crucial for reproducibility.
Experiment Setup No The paper lacks specific details regarding experimental setup, such as hyperparameter values (e.g., learning rate, batch size, number of epochs), optimizer settings, or other system-level training configurations in the main text.