List your company
ResearchFormerly WizWand

Visual Reasoning on NLVR2 v2 (dev)

88.7AccuracyX2-VLM_large
37.42850.73964.0577.361Nov 22, 2022Jan 26, 2023Apr 1, 2023Jun 5, 2023Aug 9, 2023Oct 13, 2023Dec 18, 2023
Updated 8mo ago

Evaluation results

MethodLinks
X2-VLM_large# Params=593M, Pre-training Data=More Data2022.1188.7
X2-VLM_large# Params=593M, Pre-training Data=4M2022.1187.2
X2-VLM_base# Params=255M, Pre-training Data=More Data2022.1186.2
X2-VLM_base# Params=255M, Pre-training Data=4M2022.1185.9
VLMo_large# Params=562M, Pre-training Data=4M2022.1185.6
SimVLM_large# Params=783M, Pre-training Data=More Data2022.1184.1
VLMo_base# Params=175M, Pre-training Data=4M2022.1182.8
BLIP_base# Params=240M, Pre-training Data=More Data2022.1182.5
METER# Params=341M, Pre-training Data=4M2022.1182.3
VL-BEiT# Params=175M, Pre-training Data=4M2022.1181.9
SimVLM_base# Params=273M, Pre-training Data=More Data2022.1181.7
ALBEF# Params=210M, Pre-training Data=4M2022.1180.2
CLOVAMethod Category=Tool, Backbone LLM=GPT-3.52023.1265.6
MMICLMethod Category=E2E2023.1262.2
VISPROGMethod Category=Tool, Backbone LLM=GPT-3.52023.1260.8
Visual ChatGPTMethod Category=Tool, Backbone LLM=GPT-3.52023.1251.6
OtterMethod Category=E2E2023.1248.2
GPT4TOOLSMethod Category=Tool, Backbone LLM=GPT-3.52023.1245.4
HuggingGPTMethod Category=Tool, Backbone LLM=GPT-3.52023.1244
InternGPTMethod Category=Tool, Backbone LLM=GPT-3.52023.1239.4