AO Anti-Reading Collection FT-AOs and paired taboo subjects on Qwen3-8B, used in arXiv:2607.23379. Base AO recipe: Karvonen et al. 2025 (arXiv:2512.15674). • 41 items • Updated about 16 hours ago
When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles Paper • 2607.23379 • Published Jul 25 • 18
AO Anti-Reading Collection FT-AOs and paired taboo subjects on Qwen3-8B, used in arXiv:2607.23379. Base AO recipe: Karvonen et al. 2025 (arXiv:2512.15674). • 41 items • Updated about 16 hours ago
AO Anti-Reading Collection FT-AOs and paired taboo subjects on Qwen3-8B, used in arXiv:2607.23379. Base AO recipe: Karvonen et al. 2025 (arXiv:2512.15674). • 41 items • Updated about 16 hours ago
AO Anti-Reading Collection FT-AOs and paired taboo subjects on Qwen3-8B, used in arXiv:2607.23379. Base AO recipe: Karvonen et al. 2025 (arXiv:2512.15674). • 41 items • Updated about 16 hours ago
AO Anti-Reading Collection FT-AOs and paired taboo subjects on Qwen3-8B, used in arXiv:2607.23379. Base AO recipe: Karvonen et al. 2025 (arXiv:2512.15674). • 41 items • Updated about 16 hours ago
AO Anti-Reading Collection FT-AOs and paired taboo subjects on Qwen3-8B, used in arXiv:2607.23379. Base AO recipe: Karvonen et al. 2025 (arXiv:2512.15674). • 41 items • Updated about 16 hours ago