quiz.json

{
  "lesson": "18-multilingual-nlp",
  "title": "Multilingual NLP",
  "questions": [
    {
      "stage": "pre",
      "question": "What does zero-shot cross-lingual transfer mean?",
      "options": [
        "Fine-tune a multilingual model on one source language and evaluate on a different language with no target-language labels",
        "Translating without a translation model",
        "Tokenizing with zero merges",
        "Training with zero examples"
      ],
      "correct": 0,
      "explanation": "Zero-shot transfer: train on the source language, run on the target without target-language supervision."
    },
    {
      "stage": "pre",
      "question": "Which model family ships as the standard 100-language cross-lingual baseline?",
      "options": [
        "GloVe",
        "GPT-2",
        "XLM-R (e.g. XLM-RoBERTa-base, 270M)",
        "DistilBERT"
      ],
      "correct": 2,
      "explanation": "XLM-R is the canonical 100-language pretrained baseline for cross-lingual classification."
    },
    {
      "stage": "check",
      "question": "Why does English-as-source not always give the best transfer for a non-English target?",
      "options": [
        "English has too little data",
        "Language similarity (typology, script, morphology) predicts transfer quality; a closer high-resource source can outperform English",
        "English uses BPE",
        "English is too short"
      ],
      "correct": 1,
      "explanation": "Typologically related sources (e.g. Hindi for Indic targets) often outperform English as a fine-tune source."
    },
    {
      "stage": "check",
      "question": "What is the 'fertility tax' for low-resource languages?",
      "options": [
        "BPE refuses to train",
        "Low-resource text tokenizes into more subwords per word than English, consuming context window, latency, and capacity",
        "Smaller models train slower",
        "Tokenizers cannot handle Unicode"
      ],
      "correct": 1,
      "explanation": "Long-tail languages tokenize at much higher fertility, eating context and training efficiency."
    },
    {
      "stage": "check",
      "question": "Why is per-language evaluation required, not aggregated accuracy?",
      "options": [
        "Aggregates ignore tokenization",
        "Aggregate numbers hide long-tail languages where a multilingual model can be far worse than its mean suggests",
        "Aggregates only work on classification",
        "Aggregates run faster"
      ],
      "correct": 1,
      "explanation": "Aggregate accuracy masks poor performance on low-resource languages; per-language scores expose it."
    },
    {
      "stage": "post",
      "question": "Why is fine-tuning learning rate critical when adapting a multilingual model with few-shot data?",
      "options": [
        "Lower LR wastes GPU",
        "Required by tokenizers",
        "It changes the vocabulary",
        "High LR can collapse the multilingual alignment and effectively reduce the model to English-only"
      ],
      "correct": 3,
      "explanation": "Excessive LR drifts the shared representation; conservative LR (~2e-5) preserves cross-lingual structure."
    },
    {
      "stage": "post",
      "question": "Which mitigation directly addresses tokenizer fertility for long-tail scripts?",
      "options": [
        "Skip stopwords",
        "Lower batch size",
        "Use byte-fallback (SentencePiece byte_fallback=True) or a tokenizer with broader script coverage (e.g. XLM-V)",
        "More training epochs"
      ],
      "correct": 2,
      "explanation": "Byte fallback and broader-vocab tokenizers reduce fertility and OOV for low-resource scripts."
    },
    {
      "stage": "post",
      "question": "When is a monolingual model from scratch worth trying instead of a multilingual one?",
      "options": [
        "Always for English",
        "When the target language has enough data to train a monolingual model that beats the multilingual baseline; test before assuming",
        "Only for translation",
        "Whenever the tokenizer is BPE"
      ],
      "correct": 1,
      "explanation": "Sometimes monolingual training beats multilingual for high-resource targets; empirical comparison decides."
    }
  ]
}