paper-with-me

Object Localization 벤치마크

Object Localization on IllusionVQA

9개 결과 · ⬇ CSV · JSON

Accuracy

24.3 30.65 37 43.35 49.7 2024-03 2026-09 GPT4-Vision 4-shot+CoT — 49.7 (2024-03-23) GPT4-Vision 4-shot — 46.0 (2024-03-23) Gemini-Pro — 43.5 (2024-03-23) Gemini-Pro 4-shot — 41.8 (2024-03-23) GPT4-Vision — 40.0 (2024-03-23) Gemini-Pro 4-shot+CoT — 33.9 (2024-03-23) CogVLM — 28.0 (2024-03-23) LLaVA-1.5-13B — 24.8 (2024-03-23) InstructBLIP-13B — 24.3 (2024-03-23) GPT4-Vision 4-shot+CoT — 49.7 (2024-03-23)
RankModel Accuracy PaperCodeYear
1 GPT4-Vision 4-shot+CoT 49.7 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
2 GPT4-Vision 4-shot 46 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
3 Gemini-Pro 43.5 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
4 Gemini-Pro 4-shot 41.8 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
5 GPT4-Vision 40 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
6 Gemini-Pro 4-shot+CoT 33.9 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
7 CogVLM 28 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
8 LLaVA-1.5-13B 24.8 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
9 InstructBLIP-13B 24.3 IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models csebuetnlp/illusionvqa 2024
1–9 / 9 페이지당 10 20 50 100