Our new X account is live! Follow @wizwand_team for updates

Video-to-Text Retrieval on DiDeMo

71.9R@1

InternVideo2s2-6B

Updated 2d ago

Evaluation Results

Method
InternVideo2s2-6B 2024.03	71.9	-	-	-	-	-	-
UMT-L 2024.03	65.7	-	-	-	-	-	-
UMT-L 2024.12	65.7	-	-	-	-	-	-
UMT-L 2024.12	65.7	-	-	-	-	-	-
vid-TLDR 2024.12	65.2	-	-	-	-	-	-
GRAM Model 2024.12	63.5	-	-	-	-	91.4	-
GRAM Model 2024.12	63.5	-	-	-	-	-	-
GRAM Model 2024.12	63.2	-	-	-	-	91.6	-
GRAM Model 2024.12	63.2	-	-	-	-	-	-
VAST 2024.12	62	-	-	-	-	87.7	-
VAST 2024.12	62	-	-	-	-	-	-
InternVideo-L 2024.12	59.1	-	-	-	-	-	-
VindLU 2026.01	57.2	83.1	-	-	-	88.9	-
InternVideo2-6B 2026.02	57.1	-	-	-	-	-	-
HiTeA 2024.12	56.5	-	-	-	-	-	-
VidVec 2026.02	56.5	-	-	-	-	-	-
VidVec-O 2026.02	56.5	79.7	-	-	-	86	-
InternVideo2 2025.12	54.3	-	-	-	-	-	-
Aurora 2023.05	53.1	77.4	-	1	-	85.3	-
GRAM 2024.12	52.3	-	-	-	-	-	-
GRAM Model 2024.12	52.3	-	-	-	-	80.3	-
GRAM 2024.12	52.2	-	-	-	-	-	-
GRAM Model 2024.12	52.2	-	-	-	-	78.9	-
GRAM 2026.02	52.2	-	-	-	-	-	-
UniAdapter 2023.05	52.1	77.3	-	1	-	85.2	-
UniAlign* 2026.02	51.9	-	-	-	-	-	-
PEAV-L 2025.12	51.7	-	-	-	-	-	-
UniAlign 2026.02	51.6	-	-	-	-	-	-
LoRA 2023.05	50.9	75.3	-	2	-	82.4	-
ViCLIP 2024.03	50.2	-	-	-	-	-	-
ViCLIP 2024.12	50.2	-	-	-	-	-	-
ViCLIP 2024.12	50.2	-	-	-	-	-	-
PEAV-L 2025.12	50.1	-	-	-	-	-	-
PEAV-B 2025.12	50.1	-	-	-	-	-	-
UMT-L 2024.12	49.9	-	-	-	-	-	-
UMT-L 2026.02	49.9	-	-	-	-	-	-
UMT-L 2026.02	49.9	-	-	-	-	-	-
UMT-L 2025.12	49.9	-	-	-	-	-	-
UniAdapter 2023.05	49	75.5	-	2	-	83.3	-
PEAV-S 2025.12	49	-	-	-	-	-	-
PEAV-S 2025.12	49	-	-	-	-	-	-
VAST 2024.12	48.2	-	-	-	-	-	-
VAST 2024.12	48.2	-	-	-	-	78.6	-
VAST 2026.02	48.2	-	-	-	-	-	-
MAMA 2026.01	48.1	78.1	-	-	-	85.5	-
PEAV-B 2025.12	47.8	-	-	-	-	-	-
Vid2Seq 2026.01	47.2	74.4	-	-	-	83.1	-
LamRA 2026.02	47.1	72.8	-	-	-	81.5	-
LaViLa 2026.01	47.1	74.2	-	-	-	83	-
PE-G 2025.12	46.5	-	-	-	-	-	-
CLIP-ViP 2026.01	46.3	73.2	-	-	-	81.9	-
MMRet-v1.5 2026.02	46.1	73.1	-	-	-	80.7	-
CAMoE 2021.09	45.5	71.2	-	-	-	-	-
DRL 2026.01	45.4	72.6	-	-	-	82.1	-
VIOLET 2026.01	44.8	72.9	-	-	-	82.4	-
LanguageBind 2025.12	44.7	-	-	-	-	-	-
VLM2Vec-V2 2026.02	44.2	68	-	-	-	76.1	-
CLIP4Clip 2024.03	43.6	-	-	-	-	-	-
CLIP4Clip 2024.12	43.6	-	-	-	-	-	-
CLIP4Clip 2024.12	43.6	-	-	-	-	-	-
B3 2026.02	43.3	65.3	-	-	-	72.5	-
CLIP4Clip 2023.05	42.8	68.5	-	2	-	79.2	-
VLM2Vec 2026.02	41.3	65.4	-	-	-	72.5	-
PE-L 2025.12	41	-	-	-	-	-	-
UNITE 2026.02	40.3	68.7	-	-	-	78.1	-
SigLIP2-g-opt 2025.12	40.1	-	-	-	-	-	-
LanguageBind 2023.10	39.9	66.1	-	-	-	74.6	-
LanguageBind 2024.12	39.8	-	-	-	-	-	-
LanguageBind 2024.12	39.8	-	-	-	-	76.2	-
LanguageBind 2026.02	39.8	-	-	-	-	-	-
LanguageBind 2023.10	39.7	65.5	-	-	-	73.8	-
VideoMatch 2021.10	38.7	72.9	-	2.4	-	-	-
UniME-V2 2026.02	38.7	61	-	-	-	69	-
ImageBind 2025.12	38.2	-	-	-	-	-	-
LanguageBind 2023.10	38.1	65	-	-	-	73.6	-
LanguageBind 2023.10	37.8	63.2	-	-	-	73.4	-
SigLIP2-L/16 2025.12	37.8	-	-	-	-	-	-
ALPRO 2023.05	35.9	67.5	-	3	-	78.8	-
TVTSv2 2023.10	34.6	61.9	-	-	-	71.5	-
Frozen in Time 2023.05	34.6	65	-	3	-	74.7	-
TVTSv2 2024.12	34.6	-	-	-	-	71.5	-
UMT 2023.10	34	60.4	-	-	-	68.7	-
UMT 2024.12	34	-	-	-	-	68.7	-
CLIP 2024.03	33.9	-	-	-	-	-	-
InternVideo 2022.12	33.5	-	-	-	-	-	-
InternVideo-L 2024.12	33.5	-	-	-	-	-	-
InternVideo-L 2024.12	33.5	-	-	-	-	-	-
InternVideo-L 2026.02	33.5	-	-	-	-	-	-
InternVideo-L 2026.02	33.5	-	-	-	-	-	-
OmniVL 2023.10	33.3	58.7	-	-	-	68.5	-
OmniVL 2024.12	33.3	-	-	-	-	68.5	-
All-in-one 2023.05	32.7	61.4	-	3	-	73.5	-
VIOLET 2023.05	32.6	62.8	-	-	-	74.7	-
InternVideo 2023.10	31.5	-	-	-	-	-	-
HSE 2021.10	30.1	59.2	-	3	-	-	-
ViCLIP 2024.12	27.9	-	-	-	-	-	-
ViCLIP 2024.12	27.9	-	-	-	-	-	-
ViCLIP 2026.02	27.9	-	-	-	-	-	-
ViCLIP 2026.02	27.9	-	-	-	-	-	-
UMT-L 2024.12	24.9	-	-	-	-	-	-

Showing 100 of 114 rows