ByteDance Seed, Princeton and Tsinghua propose Falcon family for training recurrent LLM memory

burkov · x · 2026-09-06

Researchers from ByteDance Seed, Princeton, Tsinghua, UCLA and Hyperbolic tackle the long-context tradeoff: full Transformer attention is expensive as sequences grow, while recurrent models keep fixed-size memory whose update rule is itself online learning. The paper argues such memory should be trained with the representation actually available at prediction time, not same-step pairing. This yields the Falcon family: normalized memory updates with explicit control over learning speed, forgetting, and per-update window size, plus chunk-parallel GPU implementations. Experiments show competitive language modeling and better extrapolation on longer arithmetic sequences.

Original post →

More from Research

Research channel →