Text-to-Video Retrieval on ACTNET-20 benchmark leaderboard · SOTA2 Research

Text-to-Video Retrieval on ACTNET-20

19.78R@1Upper Bound

Evaluation Results

Method	Links
Upper Bound2026.01		19.78	44.63	58.96	7	36.18	0.16
StructAlign2026.01		17.89	39.69	53.98	9	52.19	-0.54
StableFusion2026.01		17.71	39.4	52.76	9	62.22	0.04
VR-LwF2026.01		17.21	40.96	54.18	9	44.45	0.58
CLIP4Clip2026.01		17.07	39.96	53.43	9.47	47.38	0.45
ZSCL2026.01		16.83	38.9	52.07	9.33	65.03	0.7
X-Pool2026.01		16.57	39.83	51.82	10.22	55	0.31
TVR+CL2026.01		16.37	37.78	50.51	10.33	66.8	0.76
LwF2026.01		16.36	40.14	53.29	9.44	45.94	0.93
CLIP-ViP2026.01		16.02	37.29	50.92	10.58	48.78	0.73
MoE-Adapter2026.01		15.77	36.27	49.32	11	77.65	-0.01
ZS CLIP2026.01		14.89	34.97	47.78	12	84.02	0