Cute Four-Panel Tokenization Infographic

Originaldokument

GPT Image 2 · en

{
  "type": "Hand-drawn educational infographic illustrations",
  "style": "Adorable notebook doodle style, hand-drawn handbook aesthetics, warm beige paper background with subtle texture, playful sketch outlines, soft macaron colors, embellished with small flower and star doodles, paper card sticker layout.",
  "aspect_ratio": "Wide banner",
  "subject": {
    "characters": [
      {
        "name": "{argument name=\"dog character name\" default=\"Chai Xiaoqi\"}",
        "species": "Shiba Inu mascot",
        "appearance": "A cute orange-brown Shiba Inu with a cream-colored muzzle and belly, round face, erect ears, friendly expression, and simplified cartoon-like anatomy."
      },
      {
        "name": "{argument name=\"blue character name\" default=\"Token\"}",
        "species": "Blue Token Mascot",
        "appearance": "A rounded, triangular, blue teardrop-shaped character with a glossy, gradient body, slender limbs, rosy cheeks, a smiling expression, and the words \"...\" written on its body.{argument name=\"token label\" default=\"TOKEN\"} \""
      }
    ],
    "theme": "Comparing four text segmentation levels in a friendly, science-oriented format"
  },
  "layout": {
    "sections": [
      {
        "title": "1. Word-level segmentation",
        "position": "Leftmost panel",
        "count": 1,
        "subtitle": "Classification by complete semantics",
        "scene": "The Shiba Inu and the blue Token character stand next to a basket labeled \"Tokens,\" filled with slips of paper. Visible slips and labels read \"Programmer,\" \"Left,\" and \"Artificial Intelligence,\" while a small sign nearby reads \"Programmer Left.\" The Token character points to the basket, while the Shiba Inu displays the neatly categorized fragments."
      },
      {
        "title": "2. Character-level word segmentation",
        "position": "Left middle panel",
        "count": 1,
        "subtitle": "Text segmentation by word",
        "scene": "The blue token character is examining small character blocks scattered on the floor with a magnifying glass, while a Shiba Inu tidies them up beside them. The label at the bottom reads \"Programmer Left Artificial Intelligence,\" serving as a sample of continuous text broken down into individual characters."
      },
      {
        "title": "3. Sub-word segmentation",
        "position": "Right center panel",
        "count": 1,
        "subtitle": "Classified by word roots and prefixes/suffixes",
        "scene": "The Shiba Inu pointed at an item displaying word root breakdowns and arrows, with the blue Token mascot standing beside it. Small paper labels contained mixed fragments such as \"program\", \"person\", \"Left\", \"artificial\", and \"intelligent\", emphasizing morphemes or subword groups."
      },
      {
        "title": "4. Byte-level word segmentation",
        "position": "rightmost panel",
        "count": 1,
        "subtitle": "Convert to byte encoding combination partition",
        "scene": "The Shiba Inu uses a retro computer that spews out a long strip of printed paper, with a blue Token character standing beside it. The lower half of the paper is covered with ribbon-like strips filled with repeating byte numbers and symbols, especially \"1\", \"2\", \"3\", \"4\" and \"-3\", which are used to indicate byte encoding combinations."
      }
    ],
    "decorations": {
      "count": 10,
      "items": [
        "4 light yellow and pink tape corners",
        "Small five-pointed stars scattered around the panel",
        "Small flower graffiti between the panels",
        "Arc-shaped motion lines near the character's movements",
        "Notebook-style vertical panel dividers",
        "Rounded rectangular panel frame",
        "The light shadow behind the paper panel",
        "Faint horizontal paper texture",
        "Little Spark Icon",
        "Hand-drawn emphasis marks"
      ]
    }
  },
  "composition": "Four equal vertical panels span the entire canvas from left to right. Each panel has a prominent Chinese numerical title and a subtitle below it. Each panel features two mascots illustrating different word segmentation concepts.",
  "text": {
    "headings": [
      "1. Word-level segmentation",
      "2. Character-level word segmentation",
      "3. Sub-word segmentation",
      "4. Byte-level word segmentation"
    ],
    "subtitles": [
      "Classification by complete semantics",
      "Text segmentation by word",
      "Classified by word roots and prefixes/suffixes",
      "Convert to byte encoding combination partition"
    ]
  },
  "quality": "Clean, refined, cute, and educational visual effects suitable for presentations.",
  "use_case": "Social media science popularization: Comparing AI and NLP word segmentation methods"
}

Verwendung

1. Den vollständigen Original-Prompt einschließlich Parametern kopieren.
2. Motiv, Text und Komposition an die eigenen Anforderungen anpassen.
3. Ein kompatibles externes Werkzeug nutzen; Zugangsbedingungen und Ergebnis prüfen.

Vor dem Start

• Prüfen, ob ein Referenzbild nötig ist. Kompatibilität belegt nicht das tatsächlich verwendete Modell.
• Namen, Zahlen und Bildtexte prüfen; Modellversionen können andere Ergebnisse liefern.