Record
- Authors
- Betley, J.; et al.
- Year
- 2025
- Title
- Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
- Published in
- arXiv
- Type
- Preprint
- DOI
- 10.48550/arxiv.2512.09742
- arXiv
- arXiv:2512.09742
- Our note
- none yet
- Verification
- verified: its identifier resolves to a record with the same title and author (checked against OpenAlex, 2026-11-14).
Identifiers found: OpenAlex W4417292295 · DOI 10.48550/arxiv.2512.09742 · arXiv:2512.09742 - From the lookup
- doi (the pages' own wording is under "As cited" below)
- Id
betley-2025-weird-generalization-inductive-backdoors-new-ways
Cited on
- Automated Subtle Emergent Misalignment Research read status: not stated
- Diluting Poisoned Fine-Tuning Data Can Strengthen Its Effect read status: not stated
As cited
The pages cite this work in 2 forms.
-
Betley, J., Cocola, J., Feng, D., Chua, J., Arditi, A., Sztyber-Betley, A., and Evans, O. (2025a). Weird generalization and inductive backdoors: New ways to corrupt LLMs. arXiv:2512.09742.
-
Betley, J., et al. (2025b). Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs. arXiv:2512.09742.
BibLaTeX
@misc{betley-2025-weird-generalization-inductive-backdoors-new-ways,
author = {Betley, J. and others},
title = {{Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs}},
date = {2025},
doi = {10.48550/arxiv.2512.09742},
eprint = {2512.09742},
eprinttype = {arxiv}
}