Recursa recursively self-improving foundation model system
Recursa adapts to new domains by learning from its own experience and raises all-pass rates on held-out tasks from 30.0% to 55.0% with DeepSeek-V4.1-Flash.
2 stories, newest first · one card per story
Recursa adapts to new domains by learning from its own experience and raises all-pass rates on held-out tasks from 30.0% to 55.0% with DeepSeek-V4.1-Flash.
MoLRSI optimizes workflows and models using persistent memory to recursively improve search policies in molecular discovery.
Today 1K readers · 1.4K views · 467 reads · all time 184K views