List your company
ResearchFormerly WizWand

Visual Question Answering on MMBench (Class Accuracy)

89.97Class AccuracyGA_diff
75.222879.051482.8886.7086May 9, 2026
Updated 5mo ago

Evaluation results

MethodLinks
GA_diffBackbone=Qwen3-VL-8B, Forgetting Ratio=15%2026.0589.97
BeforeBackbone=Qwen3-VL-8B, Forgetting Ratio=5%2026.0589.95
GA_diffBackbone=Qwen3-VL-8B, Forgetting Ratio=5%2026.0589.95
BeforeBackbone=Qwen3-VL-8B, Forgetting Ratio=15%2026.0589.95
KL_MinBackbone=Qwen3-VL-8B, Forgetting Ratio=5%2026.0589.93
KL_MinBackbone=Qwen3-VL-8B, Forgetting Ratio=15%2026.0589.92
GABackbone=Qwen3-VL-8B, Forgetting Ratio=15%2026.0589.67
MMunlearnerBackbone=Qwen3-VL-8B, Forgetting Ratio=15%2026.0589.6
MMunlearnerBackbone=Qwen3-VL-8B, Forgetting Ratio=5%2026.0589.37
GABackbone=Qwen3-VL-8B, Forgetting Ratio=5%2026.0589.32
NPOBackbone=Qwen3-VL-8B, Forgetting Ratio=15%2026.0589.3
NPOBackbone=Qwen3-VL-8B, Forgetting Ratio=5%2026.0588.47
MANUBackbone=Qwen3-VL-8B, Forgetting Ratio=5%2026.0575.79
MANUBackbone=Qwen3-VL-8B, Forgetting Ratio=15%2026.0575.79