2025-12-05
@GoogleResearch It solves Retrieval, but it breaks Alignment. Titans uses “Surprise” (gradient magnitude) to decide what to write to memory. But to an AI, a typo is “surprising.” A predictable tragedy is low-gradient noise. Without a Human Anchor (\Gamma) to modulate that surprise metric, the
Google Research
Google unveils Titans, an architecture combining RNN speed with transformer accuracy for real-time learning, which can scale to context windows of 2M+ tokens
Ali Behrouz, Student Researcher, Meisam Razaviyayn, Staff Researcher, and Vahab Mirrokni, VP and Google Fellow, Google Research