[AINews] Quasi-Riemann-Hypothesis: OpenAI publishes 722 math papers solving 90 of the top 500 open math problems; “the most significant moment” in >100 years of mathematics

QF3 introduces a new off-policy reinforcement learning algorithm that trains flow policies 10x faster than previous methods for robot locomotion and manipulation tasks.

Researchers establish theoretical foundations showing how conformal prediction set sizes quantify information gain, relating them to classical information-theoretic measures.

OpenAI shares new mathematical problem-solving capabilities from an internal frontier model, including formalized proofs in Lean on GitHub.

Researchers introduce 4D-HOF, a feed-forward framework using conditional flow matching to reconstruct 4D hand-object interactions from foundation model estimates with state-of-the-art accuracy.

Researchers introduce IdeaAnchor, a training paradigm that teaches LLMs to synthesize research ideas from literature using structured specifications and retrieval.