List your company
ResearchFormerly WizWand

Meme Classification on MAMI

0.799AccuracyQwen2-VL-7B
0.489080.569540.650.73046Feb 18, 2025
Updated 8mo ago

Evaluation results

MethodLinks
Qwen2-VL-7BEvaluation Protocol=RA-HMD2025.020.7990.904
LLaVA-1.5-7BEvaluation Protocol=RA-HMD2025.020.7970.912
GPT-4o2025.020.794—
Qwen2-VL-2BEvaluation Protocol=RA-HMD2025.020.7940.893
RGCLTraining=Supervised fine-tuned2025.020.7840.894
RA-HMDbackbone=Qwen2-VL-7B, fine-tuning set=HatefulMemes, protocol=RKC2025.020.75681.4
LOREHMmode=low-resourced2025.020.754—
HateCLIPperTraining=Supervised fine-tuned2025.020.7480.872
RA-HMDbackbone=LLaVA-1.5-7B, fine-tuning set=HatefulMemes, protocol=RKC2025.020.74580
LLaVA-1.5-7BEvaluation Protocol=SFT2025.020.7390.86
Pro-Cap2025.020.7360.838
GPT-4omode=low-resourced2025.020.729—
Qwen2-VL-7BEvaluation Protocol=SFT2025.020.7240.826
RA-HMDbackbone=Qwen2-VL-2B, fine-tuning set=HatefulMemes, protocol=RKC2025.020.71174.8
Qwen2-VL-2BEvaluation Protocol=SFT2025.020.6860.777
CLIPTraining=Supervised fine-tuned2025.020.6840.777
Qwen2-VL-7BEvaluation Protocol=4-shot2025.020.6610.734
Qwen2-VL-2Bfine-tuning set=HatefulMemes, protocol=SFT + few-shot2025.020.6673.8
Qwen2-VL-7Bfine-tuning set=HatefulMemes, protocol=SFT + few-shot2025.020.6673.4
LLaVA-1.5-7BEvaluation Protocol=4-shot2025.020.6270.681
RGCLfine-tuning set=HatefulMemes2025.020.62467.8
Qwen2-VL-7Bfine-tuning set=HatefulMemes, protocol=SFT + zero-shot2025.020.61971.1
LLaVA-1.5-7Bfine-tuning set=HatefulMemes, protocol=SFT + zero-shot2025.020.61169.1
Mod-Hatemode=low-resourced2025.020.6167.4
Qwen2-VL-2BEvaluation Protocol=4-shot2025.020.5880.648
Qwen2-VL-7BEvaluation Protocol=zero-shot2025.020.5850.762
LLaVA-1.5-7BEvaluation Protocol=zero-shot2025.020.5830.676
Qwen2-VL-2Bfine-tuning set=HatefulMemes, protocol=SFT + zero-shot2025.020.57366.4
Qwen2-VL-2BEvaluation Protocol=zero-shot2025.020.510.672
LLaVA-1.5-7Bfine-tuning set=HatefulMemes, protocol=SFT + few-shot2025.020.50165.5