Open Domain Dialog On Kilt Wizard Of

KILT-F1

KILT-RL

R-Prec

ROUGE-L

Recall@5

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Vergleichstabelle

Modellname	F1	KILT-F1	KILT-RL	R-Prec	ROUGE-L	Recall@5
Modell 1	0.0	0.0	0.0	64.79	0.0	82.15
Modell 2	14.82	0.0	0.0	0.0	13.35	0.0
Modell 3	17.3	0.0	0.0	0.0	15.93	0.0
Modell 4	0.0	0.0	0.0	62.88	0.0	77.74
Modell 5	3.09	2.18	2.04	55.71	2.92	75.59
Modell 6	18.57	11.79	10.36	55.37	16.36	78.45
Modell 7	17.28	0.0	0.0	0.0	15.71	0.0
Modell 8	18.34	11.63	10.45	57.55	16.65	78.96
Modell 9	0.0	0.0	0.0	59.11	0.0	69.1
Modell 10	19.19	13.39	11.92	56.08	17.06	74.27
Modell 11	15.66	7.57	6.55	41.54	13.94	68.25
Modell 12	12.86	0.0	0.0	0.0	11.77	0.0
kilt-a-benchmark-for-knowledge-intensive	13.53	0.0	0.0	0.0	12.4	0.0
Modell 14	15.12	6.96	5.91	41.06	13.27	67.13
Modell 15	12.15	4.8	4.41	39.06	11.42	51.63
re2g-retrieve-rerank-generate-2	18.9	12.98	11.39	60.1	16.76	79.98
Modell 17	13.75	0.0	0.0	0.0	12.81	0.0
Modell 18	11.85	2.2	1.85	18.35	10.11	18.35
Modell 19	13.11	8.75	7.59	57.75	11.57	74.61
Modell 20	15.19	4.37	3.71	25.46	13.23	51.19
Modell 21	0.0	0.0	0.0	41.06	0.0	67.13