Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Puzzle Reasoning on AlgoPuzzleVQA
Loading...
28.61
Accuracy
PEPOD
21.4444
23.3047
25.165
27.0253
Mar 24, 2026
Accuracy
Updated 24d ago
Evaluation Results
Method
Method
Links
Accuracy
PEPOD
Backbone=InternVL3-2B-...
2026.03
28.61
DAPO
Backbone=InternVL3-2B-...
2026.03
27.72
PEPOG
Backbone=InternVL3-2B-...
2026.03
27.22
PEPOG
Backbone=Qwen2.5-VL-3B...
2026.03
26.94
PEPOD
Backbone=Qwen2.5-VL-3B...
2026.03
26.56
GRPO
Backbone=InternVL3-2B-...
2026.03
26.17
GRPO
Backbone=Qwen2.5-VL-3B...
2026.03
25.44
DAPO
Backbone=Qwen2.5-VL-3B...
2026.03
25.33
High-Entropy RL
Backbone=InternVL3-2B-...
2026.03
25.22
High-Entropy RL
Backbone=Qwen2.5-VL-3B...
2026.03
24.61
Base (zero-shot)
Backbone=InternVL3-2B-...
2026.03
23
Base (zero-shot)
Backbone=Qwen2.5-VL-3B...
2026.03
21.72
Feedback
Search any
task
Search any
task