List your company

Visual Puzzle Reasoning on PuzzleVQA

46.8AccuracyPEPOD
31.235.2539.343.35Mar 24, 2026

Evaluation results

MethodLinks
PEPODBackbone=Qwen2.5-VL-3B-Instruct2026.0346.8
PEPOGBackbone=InternVL3-2B-Instruct2026.0345.2
PEPODBackbone=InternVL3-2B-Instruct2026.0345.2
PEPOGBackbone=Qwen2.5-VL-3B-Instruct2026.0345
DAPOBackbone=InternVL3-2B-Instruct2026.0344.8
DAPOBackbone=Qwen2.5-VL-3B-Instruct2026.0344.6
GRPOBackbone=Qwen2.5-VL-3B-Instruct2026.0343.2
GRPOBackbone=InternVL3-2B-Instruct2026.0343.2
High-Entropy RLBackbone=Qwen2.5-VL-3B-Instruct2026.0335
High-Entropy RLBackbone=InternVL3-2B-Instruct2026.0332.6
Base (zero-shot)Backbone=Qwen2.5-VL-3B-Instruct2026.0331.8
Base (zero-shot)Backbone=InternVL3-2B-Instruct2026.0331.8