LLMが自己に関連する質問に対して「AIにすぎない」といった免責事項を追加する傾向について、チャットテンプレートの有無がその表現スタイルを切り替えるスイッチとして機能することを示す研究
StarSkirmish Bench, a benchmark where LLMs play StarCraft BW by writing code, was updated with results from flagship models of major AI companies.
The article discusses the technical challenge of enabling System 1 LLMs to exhibit System 2 competencies.
A guest post by seven mathematicians discussing the formalization of mathematics, contextualized by reactions to a recent major OpenAI release.
Sakana AI announced the release of the third generation of The AI Scientist, named 'The Agentic Scientist', on October 15, 2026.
Microsoft researchers published a paper analyzing the usage and productivity impact of command-line AI coding tools among tens of thousands of internal engineers.