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.
Influential Exemplar Replay for Incremental Learning in Recommender Systems
Authors: Xinni Zhang, Yankai Chen, Chenhao Ma, Yixiang Fang, Irwin King
AAAI 2024 | Venue PDF | LLM Run Details | Input Tokens: 19,294 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,193 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 | Extensive experiments on four prototypical backbone models, two classic recommendation tasks, and four widely used benchmarks consistently demonstrate the effectiveness of our method as well as its compatibility for extending to several incremental recommender models. |
| Researcher Affiliation | Academia | Xinni Zhang1, Yankai Chen1, Chenhao Ma2, Yixiang Fang2, Irwin King1 1The Chinese University of Hong Kong 2The Chinese University of Hong Kong, Shenzhen |
| Pseudocode | Yes | Algorithm 1: Working Procedure of INFERONCE |
| Open Source Code | No | The paper does not provide an explicit statement about the release of source code for the described methodology, nor does it include a direct link to a code repository. |
| Open Datasets | Yes | We incorporate four real-world benchmarks that vary in size, domain, sparsity, and duration, as reported in Table 1. Each dataset is partitioned chronologically into a 50% base segment and five consecutive 10% incremental segments. (Table 1 lists: Lastfm-2k, TB2014, Gowalla, Foursquare) |
| Dataset Splits | Yes | The base segment is randomly divided into training, validation, and testing sets in a 6:2:2 ratio. |
| Hardware Specification | No | The paper does not explicitly describe the specific hardware (e.g., GPU/CPU models, memory) used to run its experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., programming language versions, library versions, or solver versions). |
| Experiment Setup | Yes | Here η denotes the learning rate. Which significantly reduces the computation cost compared to Eqn. (7) whilst providing flexibility of tuning approximation rate with hyper-parameter λ. Lastly, we vary the reservoir size by adjusting the reply ratio, i.e., K/|D|, to investigate the INFERONCE performance. For stable reproducibility, we conduct five-fold cross validation. The holistic runtime costs of all models on the largest dataset Foursquare (including reservoir construction, incremental model training with early-stop, and evaluation) |