Mifano ya Adui na Maswali ya Uimara

Majibu na maelezo huonekana unapojibu maswali.

1. Under a specified threat model, what is an adversarial example?
   • An input crafted to cause a model error using changes allowed by that threat model
   • Any ordinary prediction error on unmodified input
   • An input that must contain tiny, imperceptible pixel changes
   • A model update that improves accuracy on every distribution
2. Ni nini kinachotofautisha shambulio la 'sanduku-nyeupe' na shambulio la 'sanduku-nyeusi'?
   • Mashambulizi ya sanduku nyeupe hufanya kazi kwenye picha nyeupe pekee
   • Katika kisanduku cheupe, mshambulizi anaweza kufikia vifaa vya ndani/gradient za modeli; kwenye kisanduku cheusi, matokeo pekee
   • Mashambulizi ya sanduku nyeusi huwa haraka zaidi
   • Mashambulizi ya sanduku nyeupe hayahitaji hesabu
3. FGSM (Njia ya Ishara ya Haraka ya Gradient) hutengeneza mfano usiofaa na...
   • Inachanganya pikseli bila mpangilio
   • Kuunda tena mfano kutoka mwanzo
   • Kuchukua hatua katika mwelekeo wa ishara ya gradient ya kupoteza kwa heshima na pembejeo
   • Kupunguza picha
4. What is adversarial training commonly used for?
   • Guaranteeing robustness to every possible attack
   • Proving a classifier is fair
   • Eliminating all clean-data errors
   • Improving robustness against the adversarial threat model represented during training
5. Kwa nini 'uhamisho' wa mifano ya wapinzani inahusu?
   • Mashambulizi yaliyoundwa kwa mtindo mmoja mara nyingi hupumbaza miundo mingine, na kuwezesha mashambulizi ya sanduku nyeusi
   • Inafanya mifano ya mafunzo kuwa polepole
   • Inaongeza gharama za kuhifadhi
   • Inaathiri mifano ya sauti pekee
6. How are prompt-injection attacks related to image adversarial examples?
   • They are identical bounded pixel perturbations
   • Both exploit model vulnerabilities, but prompt injection has a distinct instruction and system threat model
   • Prompt injection requires changing model weights
   • Image adversarial examples only affect text prompts
7. Why must an adversarial-robustness report specify the perturbation norm and budget?
   • They determine the model’s number of parameters
   • They guarantee transfer to every attack
   • They define which input changes the reported robustness claim covers
   • They replace the need to state an attacker’s access
8. Which pair of measures helps distinguish clean performance from performance under a specified attack?
   • Training-set size and GPU memory
   • Tokenization rate and context length
   • Model file size and parameter dtype
   • Clean accuracy and robust accuracy under the stated attack setting