Visual-Text Reference Pretraining Model for Image Captioning

Table 3

Comparison of the results generated by VTR-PTM in different input modes on MS COCO.

Approach	B@1	B@2	B@3	B@4	M	R	C	S
VTR-PTM0	71.1	55.4	40.2	29.6	24.3	51.5	100.5	20.1
VTR-PTM1	80.2	65.4	52.3	39.5	29.2	58.3	128.6	27.3
VTR-PTM2	82.9	67.3	53.4	40.9	30.9	61.5	130.2	28.5