Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Visual Understanding on ScienceQA
Loading...
89.92
Accuracy
IREASONER
87.9024
88.4262
88.95
89.4738
Jan 9, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
IREASONER
Reward Type=Continuous...
2026.01
89.92
EvoLMM
2026.01
89.5
Qwen2.5-VL-7B w/ Discrete Reward + Step-level Majority
Reward Type=Discrete,...
2026.01
88.92
Vision-Zero
external supervision=true
2026.01
88.5
Qwen2.5-VL-7B (Baseline)
Backbone=Qwen2.5-VL-7B
2026.01
88.3
Qwen2.5-VL-7B w/ Discrete Reward
Reward Type=Discrete
2026.01
87.98
Feedback
Search any
task
Search any
task