←Back
Next-generation Constitutional Classifiers: More efficient protection against universal jailbreaks
www.anthropic.com617 posts0 readers
www.anthropic.com (from sitemap)
* Latest posts *
- [SEP.09.2026]constitutional ai harmlessness from ai feedback↗
- [SEP.09.2026]measuring progress on scalable oversight for large language models↗
- [SEP.09.2026]toy models of superposition↗
- [SEP.09.2026]red teaming language models to reduce harms methods scaling behaviors and lessons learned↗
- [SEP.09.2026]language models mostly know what they know↗
- [SEP.09.2026]softmax linear units↗
- [SEP.09.2026]scaling laws and interpretability of learning from repeated data↗
- [SEP.09.2026]anthropic raises series b to build safe reliable ai↗
- [SEP.09.2026]training a helpful and harmless assistant with reinforcement learning from human feedback↗
- [SEP.09.2026]in context learning and induction heads↗
- [SEP.09.2026]predictability and surprise in large generative models↗
- [SEP.09.2026]a mathematical framework for transformer circuits↗
- [SEP.09.2026]a general language assistant as a laboratory for alignment↗
- [SEP.09.2026]anthropic raises 124 million to build more reliable general ai systems↗
- [SEP.09.2026]anthropic institute agenda↗
- [SEP.09.2026]pwc expanded partnership↗
- [SEP.09.2026]gates foundation partnership↗
- [SEP.09.2026]anthropic acquires stainless↗
- [SEP.09.2026]anthropic kpmg↗
- [SEP.09.2026]exploit evals↗