OpenCodePapers

few-shot-image-classification-on-bongard-hoi

Few-Shot Image Classification
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeAvg. AccuracyModelNameReleaseDate
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object Interactions✓ Link91.42Human (Amateur)2022-05-27
A Cognitive Paradigm Approach to Probe the Perception-Reasoning Interface in VLMs77.3GPT-4o + CA2025-01-23
Support-Set Context Matters for Bongard Problems✓ Link76.41SVM-Mimic + PMF (fine-tuned CLIP RN-50)2023-09-07
A Cognitive Paradigm Approach to Probe the Perception-Reasoning Interface in VLMs74.5Gemini 2.0 + CA2025-01-23
Support-Set Context Matters for Bongard Problems✓ Link72.45SVM-Mimic (frozen CLIP RN-50)2023-09-07
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object Interactions✓ Link55.82Meta-Baseline (ImagNet_R50)2022-05-27
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object Interactions✓ Link54.30Meta-Baseline (MoCov2_R50)2022-05-27
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object Interactions✓ Link54.23Meta-Baseline (Scratch_R50)2022-05-27
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object Interactions✓ Link49.74ANIL (ImageNet_R50)2022-05-27