Frontier Refusal-direction excision (RDE) Collection Reserved for LLM models with orthogonalized abliterations. • 6 items • Updated 5 days ago
Frontier TTS Collection Curated bleeding-edge text-to-speech and voice cloning models. • 2 items • Updated 18 days ago
Frontier Refusal-direction excision (RDE) Collection Reserved for LLM models with orthogonalized abliterations. • 6 items • Updated 5 days ago
Frontier Refusal-direction excision (RDE) Collection Reserved for LLM models with orthogonalized abliterations. • 6 items • Updated 5 days ago
Emergent behavior Collection Research regarding unintended misalignment, experimental in nature. • 2 items • Updated 27 days ago
Neural Chameleons: Language Models Can Learn to Hide Their Thoughts from Unseen Activation Monitors Paper • 2512.11949 • Published Dec 12, 2025 • 1