10 results on this page
Researchers introduce an open-source benchmark separating olympiad geometry solving from accurate diagram construction. Across 954 problems, current foundation models achieved only a 36.14% average diagram compilation success rate, revealing a substantial gap between mathematical reasoning and faithful visual construction.
FinSkillBench is a new benchmark for testing AI agents on portfolio construction, risk management, and fundamental analysis. Across nine models, curated procedural skills raised mean performance from 0.366 to 0.528, while self-generated skills delivered little benefit.
Xiaomi stabilized in Q2 2026 while expanding its AI strategy across models, chips, and robotics. Its MiMo-V2.5 model led OpenRouter call charts, the Xuanjie chip passed mass verification, and a humanoid robot began factory operations with a reported 98% success rate.
OpenAI says it has temporarily slowed scaling, paused reinforcement learning for two weeks, and placed its largest planned frontier RL run on hold while it strengthens safety evaluations. It will also offer zero data retention to eligible API customers starting in September.
OpenAI reaffirmed Zero Data Retention for eligible API customers using frontier models. It also previewed Private Safety Processing, designed to support advanced AI safety work without compromising customer data privacy.
OpenAI said it will enhance safety processes for paying users who access its most advanced AI models. The move responds to customers using AI for increasingly complex tasks and handling more sensitive information.
OpenAI has slowed parts of its AI development to strengthen security and safeguards. The company paused reinforcement-learning training for two weeks on its latest deployment-focused models and delayed its largest planned frontier RL run.
This paper frames context acquisition as active inference, enabling agents to choose among clarifying questions, retrieval, tool calls, task actions, or stopping. It introduces an Optimal Question Asking benchmark and evaluates language models on tasks involving 25 to 300 candidates, as well as clarification before generation and prompt optimization.
AWS outlines enterprise patterns for operating many agentic AI systems across diverse frameworks, models, and providers. The guidance focuses on preserving flexibility and enabling multi-agent systems to scale together without vendor lock-in.
Moonshot and ZAI are reportedly approaching the performance of models from OpenAI and Anthropic. The models can reportedly build passable websites at a cost 75% lower than Claude, highlighting potential advantages in cost-sensitive coding workflows.