Recoverable Identifier

arXiv:2604.20255 · detector doi_compliance · incontrovertible · 2026-05-20 02:08:57.327760+00:00

Paper page Integrity report arXiv Try DOI

Evidence text

Tommaso Zoppi et al. 2024. Anomaly-based Error and Intrusion Detection in Tabular Data.Future Generation Computer Systems(2024). doi:10.1016/j.future. 2024.09.002 A Appendix: Datasets Details The dataset information used in the experiments is presented in Table 3. B Appendix: Baseline Methods We compare uLEAD-TabPFN against a total of 26 baseline methods drawn from three sources. First, we include all baseline methods provided in the ADBench benchmark [18]. Second, we incorporate additional baselines evaluated by Livernoche et al. [ 32], which extend ADBench with recent deep learning-based, generative, and diffusion-based anomaly detection methods. Third, we additionally include two recently proposed methods that are not part of the above benchmarks, namely a PFN-based anomaly detection method, FoMo-0D [12], and a diffusion-based tabular anomaly detection approach, DAEE [45]. The evaluated baselines span a broad range of anomaly detection paradigms, including conventional, deep learning-based, generative, diffusion-based, and foundation model–based approaches. Specifi- cally, the conventional baselines include CBLOF [19], COPOD [30], ECOD [29], FeatureBagging [26], HBOS [14], Isolation Forest [31], kNN [40], LODA [39], LOF [4], MCD [9], OCSVM [46], and PCA [48]. Deep learning–based baselines include DeepSVDD [44] and DAGMM [58]. Following Livernoche et al. [32], we further include recent deep learning-based methods such as DROCC [16], GOAD [3], ICL [47], SLAD [55], and DIF [5

Evidence payload

{
  "printed_excerpt": "Tommaso Zoppi et al. 2024. Anomaly-based Error and Intrusion Detection in Tabular Data.Future Generation Computer Systems(2024). doi:10.1016/j.future. 2024.09.002 A Appendix: Datasets Details The dataset information used in the experiments ",
  "reconstructed_doi": "10.1016/j.future.2024.09.002",
  "ref_index": 61,
  "resolved_title": null,
  "verdict_class": "incontrovertible"
}