Lending Interaction Wings to Recommender Systems with Conversational Agents

Authors: Jiarui Jin, Xianyu Chen, Fanghua Ye, Mengyue Yang, Yue Feng, Weinan Zhang, Yong Yu, Jun Wang

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on 8 industrial datasets show that CORE could be seamlessly employed on 9 popular recommendation approaches, and can consistently bring significant improvements, compared against either recently proposed reinforcement learning-based or classical statistical methods, in both hot-start and cold-start recommendation settings.
Researcher Affiliation Academia 1Shanghai Jiao Tong University, 2University College London
Pseudocode Yes Algorithm 1 CORE for Querying Items and Attributes
Open Source Code No The paper does not include an explicit statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes We conduct experiments on 8 industrial datasets (including both tabular data, sequential behavioral data and graph-structured data) with 9 popular recommendation approaches (e.g., Deep FM [20], DIN [52]). Amazon dataset [8, 33] is a dataset collected by Amazon... Last FM dataset [9] is a dataset collected from Lastfm... Yelp dataset [12] is a dataset collected from Yelp...
Dataset Splits No The paper describes how sessions are constructed for evaluation and references a 'training set' in one table, but it does not provide specific train/validation/test dataset splits (e.g., percentages, sample counts, or citations to predefined splits) for reproducibility.
Hardware Specification Yes All the models are trained under the same hardware settings with 16-Core AMD Ryzen 9 5950X (2.194GHZ), 62.78GB RAM, NVIDIA Ge Force RTX 3080 cards.
Software Dependencies No The paper mentions following official implementations for recommendation approaches and includes a small Python snippet using 'openai' and 'os' libraries in an appendix, but it does not provide a list of specific software dependencies with version numbers for the main experimental setup.
Experiment Setup Yes The learning rate is decreased from the initial value 1 10 2 to 1 10 6 during the training process. The batch size is set as 100. The weight for the L2 regularization term is 4 10 5. The dropout rate is set as 0.5. The dimension of embedding vectors is set as 64.