RUMBA: A New Benchmark for Evaluating Long-term Memory of Dialogue Systems in Russian

RUMBA (Russian User Memory Benchmark), the first Russian-language benchmark for evaluating long-term memory in multi-session dialogues, has been introduced. It includes 85 dialogues and 1,543 questions testing information retrieval, reasoning, and the ability to refrain from answering. The benchmark accounts for temporal context and enables diagnosis of bottlenecks in memory architecture.

Сбербанк

Fields Medal Finalist Joins OpenAI: "Mathematics as a Profession May Disappear"

2026 Fields Medal laureate Jakob Zimmermann announced his transition from pure mathematics to AI safety and joining OpenAI. He believes that within a few years, AI will surpass humans in solving mathematical problems, so he will no longer take on graduate students to avoid training them for a non-existent profession.

OpenAI