Tasks
SotA
Datasets
Papers
Methods
Submit
About
SotA
/
Natural Language Processing
/
Visual Question Answering (VQA)
/
CLEVR
Visual Question Answering (VQA) on CLEVR
Metric: Accuracy (higher is better)
Leaderboard
Dataset
Loading chart...
Results
Submit a result
Export CSV
Sort:
Accuracy (best first)
Accuracy (worst first)
Date (newest first)
Date (oldest first)
Model name (A→Z)
#
Model
↕
Accuracy
▼
Extra Data
Paper
Date
↕
Code
1
NS-VQA (1K programs)
99.8
No
Neural-Symbolic VQA: Disentangling Reasoning fro...
2018-10-04
Code
2
MDETR
99.7
No
MDETR -- Modulated Detection for End-to-End Mult...
2021-04-26
Code
3
NeSyCoCo
99.7
No
NeSyCoCo: A Neuro-Symbolic Concept Composer for ...
2024-12-20
Code
4
NeSyCoCo Neuro-Symbolic
99.7
No
NeSyCoCo: A Neuro-Symbolic Concept Composer for ...
2024-12-20
Code
5
OCCAM (ours)
99.4
No
Interpretable Visual Reasoning via Induced Symbo...
2020-11-23
Code
6
TbD + reg + hres
99.1
No
Transparency by Design: Closing the Gap Between ...
2018-03-14
Code
7
NS-CL
98.9
No
The Neuro-Symbolic Concept Learner: Interpreting...
2019-04-26
Code
8
MAC
98.9
No
Compositional Attention Networks for Machine Rea...
2018-03-08
Code
9
CNN + LSTM + RN + HAN
98.8
No
Learning Visual Question Answering by Bootstrapp...
2018-08-01
Code
10
DDRprog*
98.3
No
DDRprog: A CLEVR Differentiable Dynamic Reasonin...
2018-03-30
-
11
single-hop + LCGN (ours)
97.9
No
Language-Conditioned Graph Networks for Relation...
2019-05-10
Code
12
CNN+GRU+FiLM
97.7
No
FiLM: Visual Reasoning with a General Conditioni...
2017-09-22
Code
13
XNM-Det supervised
97.7
No
Explainable and Explicit Visual Reasoning over S...
2018-12-05
Code
14
IEP-700K
96.9
No
Inferring and Executing Programs for Visual Reas...
2017-05-10
Code
15
CNN + LSTM + RN
95.5
No
A simple neural network module for relational re...
2017-06-05
Code
16
QGHC+Att+Concat
65.9
No
Question-Guided Hybrid Convolution for Visual Qu...
2018-08-08
-
#1
NS-VQA (1K programs)
SOTA
99.8
Accuracy
· 2018-10-04
Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding
Code
#2
MDETR
99.7
Accuracy
· 2021-04-26
MDETR -- Modulated Detection for End-to-End Multi-Modal Understanding
Code
#3
NeSyCoCo
99.7
Accuracy
· 2024-12-20
NeSyCoCo: A Neuro-Symbolic Concept Composer for Compositional Generalization
Code
#4
NeSyCoCo Neuro-Symbolic
99.7
Accuracy
· 2024-12-20
NeSyCoCo: A Neuro-Symbolic Concept Composer for Compositional Generalization
Code
#5
OCCAM (ours)
99.4
Accuracy
· 2020-11-23
Interpretable Visual Reasoning via Induced Symbolic Space
Code
#6
TbD + reg + hres
SOTA
99.1
Accuracy
· 2018-03-14
Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual Reasoning
Code
#7
NS-CL
98.9
Accuracy
· 2019-04-26
The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision
Code
#8
MAC
SOTA
98.9
Accuracy
· 2018-03-08
Compositional Attention Networks for Machine Reasoning
Code
#9
CNN + LSTM + RN + HAN
98.8
Accuracy
· 2018-08-01
Learning Visual Question Answering by Bootstrapping Hard Attention
Code
#10
DDRprog*
98.3
Accuracy
· 2018-03-30
DDRprog: A CLEVR Differentiable Dynamic Reasoning Programmer
#11
single-hop + LCGN (ours)
97.9
Accuracy
· 2019-05-10
Language-Conditioned Graph Networks for Relational Reasoning
Code
#12
CNN+GRU+FiLM
SOTA
97.7
Accuracy
· 2017-09-22
FiLM: Visual Reasoning with a General Conditioning Layer
Code
#13
XNM-Det supervised
97.7
Accuracy
· 2018-12-05
Explainable and Explicit Visual Reasoning over Scene Graphs
Code
#14
IEP-700K
SOTA
96.9
Accuracy
· 2017-05-10
Inferring and Executing Programs for Visual Reasoning
Code
#15
CNN + LSTM + RN
95.5
Accuracy
· 2017-06-05
A simple neural network module for relational reasoning
Code
#16
QGHC+Att+Concat
65.9
Accuracy
· 2018-08-08
Question-Guided Hybrid Convolution for Visual Question Answering