Conversational Response Selection On Ubuntu 1

R10@1

R10@2

R10@5

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Vergleichstabelle

Modellname	R10@1	R10@2	R10@5
fine-grained-post-training-for-improving	0.911	0.962	0.994
sequential-attention-based-network-for-noetic	0.796	0.894	0.975
190501969	0.882	0.949	0.990
multi-turn-response-selection-for-chatbots	0.767	0.874	0.969
global-selector-a-new-benchmark-dataset-and	0.916	0.965	0.994
response-ranking-with-multi-types-of-deep	0.821	0.911	0.981
multi-view-response-selection-for-human	0.662	0.801	0.951
modeling-multi-turn-conversation-with-deep	0.752	0.868	0.962
contextual-masked-auto-encoder-for-retrieval	0.918	0.964	0.993
the-ubuntu-dialogue-corpus-a-large-dataset-1	0.604	0.745	0.926
do-response-selection-models-really-know-what	0.875	0.942	0.988
triplenet-triple-attention-network-for-multi	0.790	0.885	0.970
sampling-matters-an-empirical-study-of	0.785	0.883	0.974
learning-an-effective-context-response	0.884	0.946	0.990
multi-hop-selector-network-for-multi-turn	0.800	0.899	0.978
sequential-matching-network-a-new	0.726	0.822	0.960
efficient-dynamic-hard-negative-sampling-for	0.917	0.965	0.994
multi-granularity-representations-of-dialog	0.753	-	-
domain-adaptive-training-bert-for-response	0.855	0.928	0.985
global-selector-a-new-benchmark-dataset-and	0.886	0.946	0.989
speaker-aware-bert-for-multi-turn-response	0.855	0.928	0.983
small-changes-make-big-differences-improving	0.886	0.948	0.990
interactive-matching-network-for-multi-turn	0.794	0.889	0.974
improved-deep-learning-baselines-for-ubuntu	0.630	0.780	0.944
one-time-of-interaction-may-not-be-enough-go	0.796	0.894	0.974