ML Research Wiki / Benchmarks / Visual Reasoning / Bongard-OpenWorld

Bongard-OpenWorld

Visual Reasoning Benchmark

Performance Over Time

📊 Showing 9 results | 📏 Metric: 2-Class Accuracy

Top Performing Models

Rank Model Paper 2-Class Accuracy Date Code
1 Gemini-2.0 + CA A Cognitive Paradigm Approach to Probe the Perception-Reasoning Interface in VLMs 93.60 2025-01-23 -
2 GPT-4o + CA A Cognitive Paradigm Approach to Probe the Perception-Reasoning Interface in VLMs 92.80 2025-01-23 -
3 Human Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World 91.00 2023-10-16 📦 joyjayng/Bongard-OpenWorld
4 SNAIL 📚 Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World 64.00 2023-10-16 📦 joyjayng/Bongard-OpenWorld
5 InstructBLIP + GPT-4 Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World 63.80 2023-10-16 📦 joyjayng/Bongard-OpenWorld
6 BLIP-2 + ChatGPT (Fine-tuned) 📚 Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World 63.30 2023-10-16 📦 joyjayng/Bongard-OpenWorld
7 InstructBLIP + ChatGPT + Neuro-Symbolic Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World 55.50 2023-10-16 📦 joyjayng/Bongard-OpenWorld
8 ChatCaptioner + ChatGPT Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World 49.30 2023-10-16 📦 joyjayng/Bongard-OpenWorld
9 Otter Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World 49.30 2023-10-16 📦 joyjayng/Bongard-OpenWorld

All Papers (9)

joyjayng/Bongard-OpenWorld
joyjayng/Bongard-OpenWorld
joyjayng/Bongard-OpenWorld
joyjayng/Bongard-OpenWorld
joyjayng/Bongard-OpenWorld
joyjayng/Bongard-OpenWorld
joyjayng/Bongard-OpenWorld