SOTAVerified

Visual Reasoning

Ability to understand actions and reasoning associated with any visual images

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 301–350 of 698 papers

TitleStatusHype
RGB-Th-Bench: A Dense benchmark for Visual-Thermal Understanding of Vision Language Models—0
Neuro-Symbolic Scene Graph Conditioning for Synthetic Image Dataset Generation—0
Chain of Functions: A Programmatic Pipeline for Fine-Grained Chart Reasoning Data—0
From Head to Tail: Towards Balanced Representation in Large Vision-Language Models through Adaptive Data Calibration—0
VERIFY: A Benchmark of Visual Explanation and Reasoning for Investigating Multimodal Reasoning Fidelity—0
SciVerse: Unveiling the Knowledge Comprehension and Visual Reasoning of LMMs on Multi-modal Scientific Problems—0
Does Acceleration Cause Hidden Instability in Vision Language Models? Uncovering Instance-Level Divergence Through a Large-Scale Empirical Study—0
Poisoned-MRAG: Knowledge Poisoning Attacks to Multimodal Retrieval Augmented Generation—0
LVLM-Compress-Bench: Benchmarking the Broader Impact of Large Vision-Language Model CompressionCode0
Towards Visual Discrimination and Reasoning of Real-World Physical Dynamics: Physics-Grounded Anomaly Detection—0
EXCLAIM: An Explainable Cross-Modal Agentic System for Misinformation Detection with Hierarchical Retrieval—0
M-LLM Based Video Frame Selection for Efficient Video Understanding—0
MMSciBench: Benchmarking Language Models on Multimodal Scientific Problems—0
Shakti-VLMs: Scalable Vision-Language Models for Enterprise AI—0
End-to-End Chart Summarization via Visual Chain-of-Thought in Vision-Language Models—0
Unraveling the geometry of visual relational reasoningCode0
Visual Reasoning Evaluation of Grok, Deepseek Janus, Gemini, Qwen, Mistral, and ChatGPT—0
VisFactor: Benchmarking Fundamental Visual Cognition in Multimodal Large Language ModelsCode0
Chitrarth: Bridging Vision and Language for a Billion People—0
KnowZRel: Common Sense Knowledge-based Zero-Shot Relationship Retrieval for Generalised Scene Graph GenerationCode0
Do we Really Need Visual Instructions? Towards Visual Instruction-Free Fine-tuning for Large Vision-Language Models—0
Learning to Stop Overthinking at Test Time—0
MET-Bench: Multimodal Entity Tracking for Evaluating the Limitations of Vision-Language and Reasoning Models—0
ZeroBench: An Impossible Visual Benchmark for Contemporary Large Multimodal Models—0
Visual Agentic AI for Spatial Reasoning with a Dynamic API—0
Boosting Multimodal Reasoning with MCTS-Automated Structured Thinking—0
Integrating LMM Planners and 3D Skill Policies for Generalizable Manipulation—0
Rethinking Bottlenecks in Safety Fine-Tuning of Vision Language Models—0
3D-MoE: A Mixture-of-Experts Multi-modal LLM for 3D Vision and Pose Diffusion via Rectified Flow—0
A Cognitive Paradigm Approach to Probe the Perception-Reasoning Interface in VLMs—0
Systematic Abductive Reasoning via Diverse Relation Representations in Vector-symbolic Architecture—0
MAPS: Advancing Multi-Modal Reasoning in Expert-Level Physical Science—0
CityLoc: 6DoF Pose Distributional Localization for Text Descriptions in Large-Scale Scenes with Gaussian Representation—0
DRIVINGVQA: Analyzing Visual Chain-of-Thought Reasoning of Vision Language Models in Real-World Scenarios with Driving Theory Tests—0
From Code to Compliance: Assessing ChatGPT's Utility in Designing an Accessible Webpage -- A Case Study—0
Socratic Questioning: Learn to Self-guide Multimodal Reasoning in the WildCode0
LogicAD: Explainable Anomaly Detection via VLM-based Text Feature Extraction—0
Language-Guided Salient Object Ranking—0
Probing Visual Language Priors in VLMs—0
Slow Perception: Let's Perceive Geometric Figures Step-by-step—0
HALLUCINOGEN: A Benchmark for Evaluating Object Hallucination in Large Visual-Language ModelsCode0
Revisiting MLLMs: An In-Depth Analysis of Image Classification Abilities—0
EarthDial: Turning Multi-sensory Earth Observations to Interactive Dialogues—0
ViUniT: Visual Unit Tests for More Robust Visual Programming—0
Progressive Multi-granular Alignments for Grounded Reasoning in Large Vision-Language ModelsCode0
MM-PoE: Multiple Choice Reasoning via. Process of Elimination using Multi-Modal ModelsCode0
Perception Tokens Enhance Visual Reasoning in Multimodal Language Models—0
VISCO: Benchmarking Fine-Grained Critique and Correction Towards Self-Improvement in Visual Reasoning—0
Learning Visual Abstract Reasoning through Dual-Stream NetworksCode0
Abductive Symbolic Solver on Abstraction and Reasoning Corpus—0
Show:102550
← PrevPage 7 of 14Next →
All datasetsWinogroundNLVR2 DevNLVR2 TestCLEVRERBongard-OpenWorldWinoGAViLVSRPHYRE-1B-CrossPHYRE-1B-WithinVASRIRFL: Image Recognition of Figurative LanguageNLVR

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1GPT-4o + CAText Score75.5—Unverified
2GPT-4V (CoT, pick b/w two options)Text Score75.25—Unverified
3GPT-4V (pick b/w two options)Text Score69.25—Unverified
4MMICL + CoCoTText Score64.25—Unverified
5GPT-4V + CoCoTText Score58.5—Unverified
6OpenFlamingo + CoCoTText Score58.25—Unverified
7GPT-4VText Score54.5—Unverified
8FIBER (EqSim)Text Score51.5—Unverified
9FIBER (finetuned, Flickr30k)Text Score51.25—Unverified
10MMICL + CCoTText Score51—Unverified
#ModelMetricClaimedVerifiedStatus
1BEiT-3Accuracy91.51—Unverified
2X2-VLM (large)Accuracy88.7—Unverified
3XFM (base)Accuracy87.6—Unverified
4X2-VLM (base)Accuracy86.2—Unverified
5CoCaAccuracy86.1—Unverified
6VLMoAccuracy85.64—Unverified
7VK-OODAccuracy84.6—Unverified
8SimVLMAccuracy84.53—Unverified
9X-VLM (base)Accuracy84.41—Unverified
10VK-OODAccuracy83.9—Unverified
#ModelMetricClaimedVerifiedStatus
1BEiT-3Accuracy92.58—Unverified
2X2-VLM (large)Accuracy89.4—Unverified
3XFM (base)Accuracy88.4—Unverified
4X2-VLM (base)Accuracy87—Unverified
5CoCaAccuracy87—Unverified
6VLMoAccuracy86.86—Unverified
7SimVLMAccuracy85.15—Unverified
8X-VLM (base)Accuracy84.76—Unverified
9BLIP-129MAccuracy83.09—Unverified
10ALBEF (14M)Accuracy82.55—Unverified
#ModelMetricClaimedVerifiedStatus
1AI CoreAverage-per ques.95.24—Unverified
2redherringAverage-per ques.91.14—Unverified
3VRDPAverage-per ques.90.24—Unverified
4FightttttAverage-per ques.88.71—Unverified
5neuralAverage-per ques.88.27—Unverified
6NERVAverage-per ques.88.05—Unverified
7DCLAverage-per ques.75.52—Unverified
8troublesolverAverage-per ques.73.3—Unverified
9v0.1Average-per ques.73.1—Unverified
10First_testAverage-per ques.69.65—Unverified
#ModelMetricClaimedVerifiedStatus
1Gemini-2.0 + CA2-Class Accuracy93.6—Unverified
2GPT-4o + CA2-Class Accuracy92.8—Unverified
3Human2-Class Accuracy91—Unverified
4SNAIL2-Class Accuracy64—Unverified
5InstructBLIP + GPT-42-Class Accuracy63.8—Unverified
6BLIP-2 + ChatGPT (Fine-tuned)2-Class Accuracy63.3—Unverified
7InstructBLIP + ChatGPT + Neuro-Symbolic2-Class Accuracy55.5—Unverified
8ChatCaptioner + ChatGPT2-Class Accuracy49.3—Unverified
9Otter2-Class Accuracy49.3—Unverified
#ModelMetricClaimedVerifiedStatus
1HumansJaccard Index90—Unverified
2ViLT (Zero-Shot)Jaccard Index52—Unverified
3X-VLM (Zero-Shot)Jaccard Index46—Unverified
4CLIP-ViT-B/32 (Zero-Shot)Jaccard Index41—Unverified
5CLIP-ViT-L/14 (Zero-Shot)Jaccard Index40—Unverified
6CLIP-RN50x64/14 (Zero-Shot)Jaccard Index38—Unverified
7CLIP-RN50 (Zero-Shot)Jaccard Index35—Unverified
8CLIP-ViL (Zero-Shot)Jaccard Index15—Unverified
#ModelMetricClaimedVerifiedStatus
1LXMERTaccuracy70.1—Unverified
2ViLTaccuracy69.3—Unverified
3CLIP (finetuned)accuracy65.1—Unverified
4CLIP (frozen)accuracy56—Unverified
5VisualBERTaccuracy55.2—Unverified
#ModelMetricClaimedVerifiedStatus
1RPINAUCCESS42.2—Unverified
2Dec[Joint]1fAUCCESS40.3—Unverified
3Dynamics-Aware DQNAUCCESS39.9—Unverified
4DQNAUCCESS36.8—Unverified
#ModelMetricClaimedVerifiedStatus
1RPINAUCCESS85.2—Unverified
2Dynamics-Aware DQNAUCCESS85.2—Unverified
3Dec[Joint]1fAUCCESS80—Unverified
4DQNAUCCESS77.6—Unverified
#ModelMetricClaimedVerifiedStatus
1Swin1:1 Accuracy52.9—Unverified
2ConvNeXt1:1 Accuracy51.2—Unverified
3ViT1:1 Accuracy50.3—Unverified
4DEiT1:1 Accuracy47.2—Unverified
#ModelMetricClaimedVerifiedStatus
1Humans1-of-100 Accuracy100—Unverified
#ModelMetricClaimedVerifiedStatus
1VisualBERTAccuracy (Dev)67.4—Unverified