Reranking Models Quiz

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

1. Mupombi yekudzoreredza yematanho maviri, basa reranker nderei?
   • Kupa zvakare zvibodzwa uye kurongazve rondedzero pfupi yevachakwikwidza kuti ienderane
   • Kudhonza vamiriri vekutanga kubva kumamirioni ezvinyorwa
   • Kuchengeta magwaro mudhatabhesi
   • Kugadzira mhinduro yekupedzisira yemutauro wechisikigo
2. Besides MS MARCO, on which dataset did Nogueira and Cho report strong results for their BERT passage reranker?
   • ImageNet
   • TREC-CAR
   • LibriSpeech
   • SQuAD only
3. Sei muchinjika-encoder reranker isinga shandiswe kutsvaga mamirioni ezvinyorwa zvakananga?
   • Inoshanda chete pamifananidzo
   • Haikwanise kuverenga zvinyorwa
   • Zvibodzwa zvaro zvinofanirwa kuverengerwa zvinyowani kune yega yega yemubvunzo-gwaro peya, iyo inononoka pachiyero.
   • Hazvidi computation zvachose
4. Ndeipi mukana mukuru webi-encoder pamusoro pemuchinjiko-encoder?
   • Yakanyanya kujeka pakukoshera kwakaringana
   • Inogara ichirongedza mhedzisiro zvakakwana
   • Izvo hazvidi data yekudzidziswa
   • Kumisikidzwa kwegwaro rayo kunogona kufanoverengerwa, zvichiita kuti kudzosa nekukurumidza uye kuwedzere
5. How can a reranker affect context quality in a retrieval-augmented generation pipeline?
   • It can move more relevant retrieved candidates higher in the context shortlist, but does not guarantee correctness
   • It ensures every retrieved passage is factually correct
   • It creates missing documents in the corpus
   • It removes the need to evaluate retrieval
6. In listwise reranking, what does the model consider together?
   • Only one document with no query
   • A query and a set of candidate documents when deciding their ordering
   • Each document embedding before any candidate retrieval
   • A single relevance label for the full training corpus
7. What should a team verify before interpreting reranker scores as probabilities?
   • Whether every score is above zero
   • Whether the first-stage retriever was removed
   • Whether the scores have been calibrated and evaluated for that use
   • Whether the model has more parameters than the retriever
8. What limitation does a reranker inherit from the first-stage retrieval system?
   • It always returns a calibrated probability
   • It removes the need for retrieval
   • It guarantees a correct generated answer
   • It cannot promote a relevant document that was never included in the candidate set