Guide-yakabatana mibvunzo · Pakati Nhanho

Prompt Tuning Quiz

Ongorora zvinyoro zvinyoro pane yekuisa layer, iyo yakaomeswa-modhi setup, uye chiyero-inotsamira mhedzisiro yakashumwa kuti igadzirise nekukurumidza.

Nzira dzekutungamira dzinoenderanaPrompt Tuning
Mubvunzo 1 ye 8

Chii chinonzi 'soft prompt' mukugadzirisa nekukasira?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Chii chinonzi 'soft prompt' mukugadzirisa nekukasira?
    • Akadzidza kuenderera mberi nekumisikidza mavheta asingade mepu kumashoko chaiwo
    • Murairo wemashoko ane ruremekedzo
    • Ikopi diki yakasvibiswa yemuenzaniso
    • Seti yemarogi akaomeswa
  2. Ndekupi kukurumidza tuning kunowedzera mavheta ayo anodzidziswa?
    • Pane yega yega transformer layer
    • Chete pane embedding layer
    • Chete pakubuda layer
    • Mukati meiyo optimizer state
  3. Ndeapi mhedzisiro inokatyamadza iyo bepa rekutanga rakasimbisa pamusoro pechiyero chemuenzaniso?
    • Kurumidza kugadzirisa kunowedzera kuipa sezvo modhi dzinokura
    • Chikero hachina mhedzisiro pakugadzirisa nekukasira
    • Sezvo mamodheru achikura, kugadzirisa nekukasira kunosvika kuzara-tuning mhando
    • Mamodheru akakura haachagone kugadziridzwa nekukurumidza
  4. Ndiani akatanga tuning yekukurumidza?
    • DeepMind muna 2016
    • Anthropic muna 2023
    • Microsoft muna 2019
    • Vatsvagiri vepa Google (Lester, Al-Rfou, Constant) muna 2021
  5. Ko kukurumidza tuning inogonesa sei kushumira akawanda mabasa kubva kune imwe modhi?
    • Basa rega rega rinoshandisa rayo diki nyoro kukurumidza apo modhi inogara yakaoma nechando
    • Inodzokorora modhi yose yechikumbiro chega chega
    • Inobvisa mabasa asina kushandiswa kubva kune uremu
    • Inoda GPU yakasiyana nebasa
  6. How does Lester and colleagues’ paper describe prompt tuning in relation to prefix tuning?
    • As a replacement for frozen language models
    • As a simpler related method that the paper compares with prefix tuning
    • As a tokenizer with no trainable parameters
    • As a method that updates every model weight
  7. Which model family did Lester and colleagues use to study prompt tuning across scales?
    • WaveNet
    • AlphaFold2
    • T5
    • Isolation Sango
  8. How are soft prompts optimized in prompt tuning?
    • By changing all model weights for each task
    • By replacing embeddings with hard-coded words only
    • By training a new tokenizer for every example
    • By backpropagation through a frozen language model to update the prompt vectors