Efficient Dynamic Hard Negative Sampling for Dialogue Selection
Open paper
Benchmark
Model
Rank
Results
conversational-response-selection-on-e
BERT-FP+EDHNS
#1
R10@1: 0.957
R10@2: 0.986
R10@5: 0.997
conversational-response-selection-on-ubuntu-1
BERT-FP+EDHNS
#2
R10@1: 0.917
R10@2: 0.965
R10@5: 0.994
Rank counts only results with a code link.