← Multimodal Sentiment Analysis

Papers with Code · CC BY-SA 4.0

CMU-MOSEI

아카이브 평가 15건 · 마지막 적재/변경 2026-10-30

원본 데이터셋 링크 1

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
SeMUL-PCD	Accuracy 88.62 F1 89.04	Multi-label Emotion Analysis in Conversation via Multimodal Knowledge Distillation 2023-10-27	연결 없음
MMML	Accuracy 88.22 MAE 0.5072 F1 88.04 Acc-7 54.77 Acc-5 57.45 Corr 81.42	Multimodal Multi-loss Fusion Network for Sentiment Analysis 2023-08-01	코드 1
UniMSE	Accuracy 87.50 MAE 0.523 F1 87.46	UniMSE: Towards Unified Multimodal Sentiment Analysis and Emotion Recognition 2022-11-21	코드 1
SPECTRA	Accuracy 87.34	Speech-Text Dialog Pre-training for Spoken Dialog Understanding with Explicit Cross-Modal Alignment 2023-05-19	코드 1
Transformer-based joint-encoding	Accuracy 82.48	A Transformer-based joint-encoding for Emotion Recognition and Sentiment Analysis 2020-06-29	코드 1
Modulated-fusion transformer	Accuracy 82.45	Modulated Fusion using Transformer for Linguistic-Acoustic Emotion Recognition 2020-10-05	코드 1
MMLatch	Accuracy 82.4 MAE 0.7	MMLatch: Bottom-up Top-down Fusion for Multimodal Sentiment Analysis 2022-01-24	코드 1
Multilogue-Net	Accuracy 82.10 MAE 0.59	Multilogue-Net: A Context Aware RNN for Multi-modal Emotion Detection and Sentiment Analysis in Conversation 2020-02-19	코드 1
Proposed: B2 + B4 w/ multimodal fusion	Accuracy 81.14	Gated Mechanism for Attention Based Multimodal Sentiment Analysis 2020-02-21	연결 없음
CAE-LR	Accuracy 78	Unsupervised Multimodal Language Representations using Convolutional Autoencoders 2021-10-06	연결 없음
Graph-MFN	Accuracy 76.9 MAE 0.71	Multimodal Language Analysis in the Wild: CMU-MOSEI Dataset and Interpretable Dynamic Fusion Graph 2018-07-01	연결 없음
MARLIN (ViT-L)	Accuracy 74.83	MARLIN: Masked Autoencoder for facial video Representation LearnINg 2022-11-12	코드 1
MARLIN (ViT-B)	Accuracy 73.7	MARLIN: Masked Autoencoder for facial video Representation LearnINg 2022-11-12	코드 1
MARLIN (ViT-S)	Accuracy 72.69	MARLIN: Masked Autoencoder for facial video Representation LearnINg 2022-11-12	코드 1
ALMT	MAE 0.526 Acc-7 54.28 Acc-5 55.96 Corr 0.779	Learning Language-guided Adaptive Hyper-modality Representation for Multimodal Sentiment Analysis 2023-10-09	코드 1