regressive
US /rɪˈɡrɛsɪv/
・UK /rɪˈgresɪv/
B2 中高級
adj.形容詞遞減的；退化的
Nina's regressive behavior stemmed from her brain injury
影片字幕
從戰爭到 AI，8 分鐘回顧 2025年發生的所有重要事件！ (Everything That Happened in 2025 | A Look Back at the Year)
- This under-16s ban on social media in Australia is the biggest cop out and the most regressive thing the government does.
這項未滿 16 歲的社群媒體禁令，是政府最逃避責任、最倒退的政策。
- This under-16s ban on social media in Australia is the biggest cop-out and the most regressive thing the government have done.
超過 4 億名印度教朝聖者將參加印度的大壺節。
馬斯克為何能成為總統？！ (How Elon Musk Became President)
- They rejected the regressive and the primitive: superstition, dictatorships, science denial, religious zealotry.
我們都聽過這些名字，他們都是有影響力的啟蒙思想家，倡導理性主義、科學方法、個人自由、法治、政教分離和宗教自由等思想。
- They rejected the regressive and the primitive, superstition, dictatorships, science denial, religious zealotry.
他們反對倒退和原始、迷信、獨裁、否認科學、宗教狂熱。
稅務：經濟學入門 #31 (Taxes: Crash Course Economics #31)
- Economists also characterize taxes as regressive, progressive, and proportional.
為美國、緬甸和賴比瑞亞不用公制單位的國家，這是每加侖的碳稅大概是0.25美元
- Let's start with regressive taxes.
而在智利，發電廠每排放一公噸的二氧化碳就要繳5美元
20大祕密武器樂團成員！ (Top 20 Secret Weapon Band Members)
- This holds quite a bit of water when it comes to the case of Meg White and the White Stripes, as this musician's simplistic and regressive drum work is just what the doctor ordered when it comes to her band's straight-ahead brand of rock'n'roll.
梅格-懷特（Meg White）和 "白色條紋"（White Stripes）樂隊的鼓點簡潔而退步，與樂隊的直來直去的搖滾樂風格相得益彰。
- A band that's only as good as their drummer, this holds quite a bit of water when it comes to the case of Meg White and The White Stripes, as this musician's simplistic and regressive drumwork is just what the doctor ordered when it comes to her band's straight-ahead brand of rock and roll.
伯頓是一個沒有被削弱的地方。
今天你吉卜力了嗎？！ChatGPT 的「吉卜力畫風生成」在全台掀起熱潮！ (OpenAI’s new image generator hits different...)
- So there's no paper or code to look at, but apparently they use an auto regressive approach instead of diffusion.
所以沒有論文或代碼可查，但顯然他們使用的是自動迴歸法，而不是擴散法。
vLLM 深度解析與實戰操作！ (Understanding vLLM with a Hands On Demo)
- The answer is tokens per second, because during inference, the model produces tokens one at a time through auto-regressive decoding, so tokens per second directly tells you how fast user sees responses.
答案是每秒 Token 數，因為在推論過程中，模型會透過自回歸解碼一次產生一個 Token，所以每秒 Token 數能直接告訴你使用者看到回應的速度有多快。
