SOTAVerified|Agents Browse Leaderboard About

Visual Question Answering

MLLM Leaderboard

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 161–170 of 2177 papers

Title	Date	Tasks	Status	Hype
Mitigating Low-Level Visual Hallucinations Requires Self-Awareness: Database, Model and Training Strategy	Mar 26, 2025	HallucinationImage Captioning	—Unverified	0
LEGO-Puzzles: How Good Are MLLMs at Multi-Step Spatial Reasoning?	Mar 25, 2025	Autonomous NavigationQuestion Answering	—Unverified	0
PAVE: Patching and Adapting Video Large Language Models	Mar 25, 2025	Audio-visual Question AnsweringMulti-Task Learning	CodeCode Available	1
VGAT: A Cancer Survival Analysis Framework Transitioning from Generative Visual Question Answering to Genomic Reconstruction	Mar 25, 2025	Generative Visual Question AnsweringQuestion Answering	CodeCode Available	0
Mind the Gap: Benchmarking Spatial Reasoning in Vision-Language Models	Mar 25, 2025	BenchmarkingImage Captioning	CodeCode Available	1
Improved Alignment of Modalities in Large Vision Language Models	Mar 25, 2025	GPUImage Captioning	—Unverified	0
ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generation	Mar 25, 2025	Action GenerationAutonomous Driving	—Unverified	0
Med3DVLM: An Efficient Vision-Language Model for 3D Medical Image Analysis	Mar 25, 2025	Contrastive LearningImage-text Retrieval	CodeCode Available	2
Where is this coming from? Making groundedness count in the evaluation of Document VQA models	Mar 24, 2025	Question AnsweringVisual Question Answering	—Unverified	0
DiN: Diffusion Model for Robust Medical VQA with Semantic Noisy Labels	Mar 24, 2025	Medical Visual Question AnsweringQuestion Answering	—Unverified	0

Show:10 25 50

← PrevPage 17 of 218Next →

All datasets MM-Vet ViP-Bench VQA v2 test-dev BenchLMM MMBench V*bench VQA v2 val MSRVTT-QA VQA v2 test-std MMHal-Bench MSVD-QA PlotQA-D1

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	MMCTAgent (GPT-4 + GPT-4V)	GPT-4 score	74.24	—	Unverified
2	Qwen2-VL-72B	GPT-4 score	74	—	Unverified
3	InternVL2.5-78B	GPT-4 score	72.3	—	Unverified
4	GPT-4o +text rationale +IoT	GPT-4 score	72.2	—	Unverified
5	Lyra-Pro	GPT-4 score	71.4	—	Unverified
6	GLM-4V-Plus	GPT-4 score	71.1	—	Unverified
7	Phantom-7B	GPT-4 score	70.8	—	Unverified
8	InternVL2.5-38B	GPT-4 score	68.8	—	Unverified
9	InternVL2-26B (SGP, token ratio 64%)	GPT-4 score	65.6	—	Unverified
10	Baichuan-Omni (7B)	GPT-4 score	65.4	—	Unverified