Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Vision-Language Modeling on CharXiv
Loading...
73.9
Accuracy
ZPPO
53.516
58.808
64.1
69.392
Jun 16, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
ZPPO
Model Scale=2B, Prompt...
2026.06
73.9
GRPO
Model Scale=2B, Prompt...
2026.06
73
GRPO
Model Scale=2B, Prompt...
2026.06
72.8
On-Distill
Model Scale=2B, Prompt...
2026.06
71.7
On-Distill
Model Scale=2B, Prompt...
2026.06
71.7
Qwen3.5
Model Scale=2B, Prompt...
2026.06
71.6
Off-Distill
Model Scale=2B, Prompt...
2026.06
71.4
Off-Distill
Model Scale=2B, Prompt...
2026.06
71
ZPPO
Model Scale=0.8B, Prom...
2026.06
63.9
GRPO
Model Scale=0.8B, Prom...
2026.06
59.9
GRPO
Model Scale=0.8B, Prom...
2026.06
57
On-Distill
Model Scale=0.8B, Prom...
2026.06
56.4
Off-Distill
Model Scale=0.8B, Prom...
2026.06
56
On-Distill
Model Scale=0.8B, Prom...
2026.06
56
Off-Distill
Model Scale=0.8B, Prom...
2026.06
55.1
Qwen3.5
Model Scale=0.8B, Prom...
2026.06
54.3
Feedback
Search any
task
Search any
task