Полковник высказался о новом уровне конфликта Ирана с США и Израилем14:52
Гуменник рассказал о переживаниях перед финалом Гран-при России17:42
。wps对此有专业解读
My first instinct was creativity. I had models generate poems, short stories, metaphors, the kind of rich, open-ended output that feels like it should reveal deep differences in cognitive ability. I used an LLM-as-judge to score the outputs, but the results were pretty bad. I managed to fix LLM-as-Judge with some engineering, and the scoring system turned out to be useful later for other things, so here it is:
Trust plays a decisive role. High-trust environments reduce coordination costs and accelerate execution. When monitoring feels transparent and supportive, adoption tends to follow. When it feels extractive, stress responses increase and intrinsic motivation declines. Surveillance may increase visible output in the short term, but it can quietly raise the long-term cost structure of the organization.