引導式測驗 · 中等 等級

穩定性人工智慧測驗

回顧 Stability AI 的 Stable Diffusion 版本、潛在擴散方法、訓練資料和選定的音訊產生工作。

相關引導路徑Stability AI
問題 1 的 8

Stability AI 在 2022 年發布的最出名的產品是什麼?

預覽每個問題8 問題

當您參加測驗時,就會出現答案和解釋。

  1. Stability AI 在 2022 年發布的最出名的產品是什麼?
    • 開放權重影像產生器穩定擴散
    • 一個名為 Claude 的封閉式聊天機器人
    • 自動駕駛汽車平台
    • 專有搜尋引擎
  2. Stability AI 在 2022 年 Stable Diffusion 發布時公開了什麼?
    • Unrestricted public-domain weights for every model
    • Model weights downloadable under that release’s license terms
    • The weights of DALL-E and Midjourney
    • A license allowing any use without conditions
  3. 從技術上講,穩定擴散在哪裡執行其去雜訊過程?
    • 直接在全解析度像素上
    • 區塊鏈分類帳內部
    • 在自動編碼器產生的壓縮潛在空間中
    • 僅適用於音訊波形
  4. 根據 Stable Diffusion v1-4 模型卡,該檢查點使用了哪些訓練資料?
    • LAION-2B(en) and subsets, including a fine-tuning stage on LAION-Aesthetics v2 5+
    • A private catalog of licensed studio images only
    • The full LAION-5B dataset with no filtering or subsets
    • A collection of audio-transcript pairs
  5. 除了影像之外,Stability AI 還為哪種模式建立模型?
    • 透過穩定音訊生成音頻
    • 僅電子表格
    • 稅務準備
    • 僅物理機器人硬體
  6. In Getty Images v Stability AI, what happened to the primary training/development and outputs claims at trial?
    • The court held all model training lawful worldwide
    • They were abandoned at trial; the secondary copyright-infringement claim was dismissed
    • The court found infringement for every generated image
    • The court dismissed every claim because Getty lacked copyright
  7. What kind of data did Stability AI describe for Stable Audio Open?
    • Only private voice recordings from customers
    • Only MIDI scores without audio
    • Creative Commons-licensed audio data
    • A dataset of Stable Diffusion image prompts
  8. What release format did Stability AI announce for Stable Audio 3.0?
    • A closed image-only classifier
    • A text-only language model with no audio output
    • A codec that accepts no learned weights
    • Open-weight music models trained on fully licensed data