Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates.
RRL: Recommendation Reverse Learning
Authors: Xiaoyu You, Jianwei Xu, Mi Zhang, Zechen Gao, Min Yang
AAAI 2024 | Venue PDF | LLM Run Details | Input Tokens: 17,572 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 6,891 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct experiments on two representative recommendation models and three public benchmark datasets to verify the efficiency of RRL. |
| Researcher Affiliation | Academia | Xiaoyu You, Jianwei Xu, Mi Zhang*, Zechen Gao, Min Yang* Fudan University EMAIL, EMAIL, mi EMAIL, EMAIL, m EMAIL, |
| Pseudocode | No | The paper provides mathematical equations for its framework but does not include structured pseudocode or an algorithm block. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. |
| Open Datasets | Yes | We use three real-world recommendation datasets, i.e., Gowalla (Liang et al. 2016), Yelp2018 (Wang et al. 2019), and Movilens-1m (dubbed as ML-1M)3 which are widely used for benchmarking. 3https://grouplens.org/datasets/movielens/1m/ |
| Dataset Splits | Yes | Each dataset is split into training/validation/testing sets by the ratio of 70/10/20%. Validation sets are used to tune hyper-parameters. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types, or memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment. |
| Experiment Setup | Yes | The depth Light GCN is set to 3, and the size of the embeddings of recommendation models is set as 128. For each baseline model, we mostly follow the suggested experimental settings and set the hyper-parameters as suggested in the original papers, including but not limited to the learning rate, the regularization coefficient, etc. For shilling attacks used to verify removal completeness, we randomly choose an unpopular item as the target item to measure #Rec, and the number of malicious users is set as 10% of all users. |