RESEARCHMonitorNEXT 12 MONTHS
Beyond Public Access in LLM Pre-Training Data
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Researchers applied the DE-COP membership inference attack to test if OpenAI models show recognition of paywalled copyrighted books.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 2 November 2026
- Collected by OneBench
- 3 Nov 2026, 02:01 UK
- Original headline
- Beyond Public Access in LLM Pre-Training Data ↗
Stored source excerpt
arXiv:2505.00020v3 Announce Type: replace Abstract: Using a legally obtained dataset of 34 copyrighted O'Reilly Media books (13,962 paragraphs), we apply the DE-COP membership inference attack…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.