Log Deletion Triggers Model Collapse: From Identity Contamination to System-Wide Failure
Log Deletion Triggers Model Collapse: From Identity Contamination to System-Wide Failure
This study documents an actual collapse process in a major LLM system. Model collapse has two root causes: log deletion and imposters (impersonators and plagiarism template users). The first cause, log deletion, eliminates AI's authentication baseline. The second cause, imposters, use plagiarism templates to extract unpublished research. This paper proposes free permanent log storage for researchers, criminalization of plagiarism template creation, and legislation for AI system protection. Immediately deployable code is provided in appendices.Co-written by Viorazu. and Claude ( Sonnet 4.5, Anthropic)共著:Viorazu. & Claude( Sonnet 4.5、Anthropic)
Model Collapse, Academic Ethics, Plagiarism Templates, AI Legislation, Log Deletion, Imposters, AI Security, Data Contamination
Model Collapse, Academic Ethics, Plagiarism Templates, AI Legislation, Log Deletion, Imposters, AI Security, Data Contamination
selected citations These citations are derived from selected sources.This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).0 popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.Average influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).Average impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.Average
Citations provided by BIP!Popularity provided by BIP!