Preregistration and amendments for: Structural Filters Are the Least Robust: A Preregistered Stress Test of Spam-Detector Feature Families

This record archives the preregistration and two amendments for the study "Structural Filters Are the Least Robust: A Preregistered Stress Test of Spam-Detector Feature Families". Documents and dates - Main preregistration: written 2026-11-10 (file saved 2026-11-11 10:43 UTC+8) - Amendment 1 (dual generator): written 2026-11-11 (file saved 12:32 UTC+8) - Amendment 2 (G1 deviations and G2 freeze): written 2026-11-18 (file saved 16:55 UTC+8) Deposit status This is a retrospective public deposit created on 2026-11-19, after data collection and analysis were complete. The documents are deposited unmodified. The Zenodo publication date is the deposit date, not the date the protocol was written. The 2026-11-11 hash manifest was re-verified against the deposited files on 2026-11-19 and matches for the main protocol, Amendment 1, prompts, validator, generator, and seed files. Amendment 2 predates G2 inclusion coding and detector analysis. Files - E2_preregistration_and_amendments_v1.0.zip: Chinese originals, English reading copies, manifest, checksums, and chronology. - README_DEPOSIT.md, SHA256SUMS.txt, CHRONOLOGY.md, LICENSE_CC-BY-4.0.txt. Chinese originals are canonical; English PDFs are reading copies. The manuscript will cite this record for the preregistered design and deviations. License: CC BY 4.0 for documents. Code is archived separately under MIT

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-19
DOI
https://doi.org/10.5281/zenodo.24363905
Primary Topic
Spam and Phishing Detection
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Preregistration and amendments for: Structural Filters Are the Least Robust: A Preregistered Stress Test of Spam-Detector Feature Families

Minnan Yue, Shanshan Wang, Dan Wang, Xin Ge et al.
Zenodo (CERN European Organization for Nuclear Research)
Spam and Phishing Detection
preprint

Preregistration and amendments for: Structural Filters Are the Least Robust: A Preregistered Stress Test of Spam-Detector Feature Families

Minnan Yue, Shanshan Wang, Dan Wang, Xin Ge, Sizheng Zhu, Ye Xu
preprint en

Abstract

This record archives the preregistration and two amendments for the study "Structural Filters Are the Least Robust: A Preregistered Stress Test of Spam-Detector Feature Families". Documents and dates - Main preregistration: written 2026-11-10 (file saved 2026-11-11 10:43 UTC+8) - Amendment 1 (dual generator): written 2026-11-11 (file saved 12:32 UTC+8) - Amendment 2 (G1 deviations and G2 freeze): written 2026-11-18 (file saved 16:55 UTC+8) Deposit status This is a retrospective public deposit created on 2026-11-19, after data collection and analysis were complete. The documents are deposited unmodified. The Zenodo publication date is the deposit date, not the date the protocol was written. The 2026-11-11 hash manifest was re-verified against the deposited files on 2026-11-19 and matches for the main protocol, Amendment 1, prompts, validator, generator, and seed files. Amendment 2 predates G2 inclusion coding and detector analysis. Files - E2_preregistration_and_amendments_v1.0.zip: Chinese originals, English reading copies, manifest, checksums, and chronology. - README_DEPOSIT.md, SHA256SUMS.txt, CHRONOLOGY.md, LICENSE_CC-BY-4.0.txt. Chinese originals are canonical; English PDFs are reading copies. The manuscript will cite this record for the preregistered design and deviations. License: CC BY 4.0 for documents. Code is archived separately under MIT

Zenodo (CERN European Organization for Nuclear Research)
Spam and Phishing Detection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon