CNN+LSTM

Reported on 4 benchmarks across 4 tasks · 3 papers · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing2 results

Visual Question Answering (VQA)onGQA test-std
Accuracy· 2019-02-25
46.55
best: 65.14 (ProTo)
GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering arXiv:1902.09506
Sentiment AnalysisonIMDb
Accuracy· uses extra data· 2017-07-06
88.9
best: 96.68 (RoBERTa-large with LlamBERT)
On the Role of Text Preprocessing in Neural Network Architectures: An Evaluation Study on Text Categorization and Sentiment Analysis arXiv:1707.01780

Audio1 result

Emotion RecognitiononIEMOCAP
UA· 2018-02-15
0.65
best: 0.74 (SYSCOMB: BLSTMATT with CSA (session5))
SOTA
CNN+LSTM Architecture for Speech Emotion Recognition with Data Augmentation arXiv:1802.05630

Speech1 result

Speech Emotion RecognitiononIEMOCAP
UA· 2018-02-15
0.65
best: 0.74 (SYSCOMB: BLSTMATT with CSA (session5))
SOTA
CNN+LSTM Architecture for Speech Emotion Recognition with Data Augmentation arXiv:1802.05630