AAAI 2020 accepted papers on Vision-language and multimodal
60 of the 1,610 papers accepted at AAAI 2020 (3.7%) are on Vision-language and multimodal, compared with 4.4% at AAAI 2019. They fall into 5 topics.
See AAAI 2020 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Visual question answering · 20 papers
- Reasoning with Heterogeneous Graph Alignment for Video Question Answering (195 citations)
- Federated Learning for Vision-and-Language Grounding Problems (118 citations)
- Overcoming Language Priors in VQA via Decomposed Linguistic Representations (103 citations)
- Re-Attention for Visual Question Answering (98 citations)
- Getting Closer to AI Complete Question Answering: A Set of Prerequisite Real Tasks (74 citations)
- DualVD: An Adaptive Dual Encoding Model for Deep Visual Understanding in Visual Dialogue (70 citations)
- Joint Learning of Answer Selection and Answer Summary Generation in Community Question Answering (56 citations)
- Improving Knowledge-Aware Dialogue Generation via Knowledge Base Question Answering (47 citations)
- ManyModalQA: Modality Disambiguation and QA over Diverse Inputs (40 citations)
- Visual Dialogue State Tracking for Question Generation (33 citations)
- Explanation vs Attention: A Two-Player Game to Obtain Attention for VQA (28 citations)
- DMRM: A Dual-Channel Multi-Hop Reasoning Model for Visual Dialog (27 citations)
and 8 more in the interactive map.
Human and embodied modeling · 20 papers
- Unified Vision-Language Pre-Training for Image Captioning and VQA (896 citations)
- Unicoder-VL: A Universal Encoder for Vision and Language by Cross-Modal Pre-Training (821 citations)
- When Radiology Report Generation Meets Knowledge Graph (363 citations)
- Decoupled Attention Network for Text Recognition (318 citations)
- Pose-Guided Multi-Granularity Attention Network for Text-Based Person Search (205 citations)
- Learning Cross-Modal Context Graph for Visual Grounding (85 citations)
- MemCap: Memorizing Style Knowledge for Image Captioning (75 citations)
- Expressing Objects Just Like Words: Recurrent Visual Embedding for Image-Text Matching (70 citations)
- Aspect-Aware Multimodal Summarization for Chinese E-Commerce Products (66 citations)
- Show, Recall, and Tell: Image Captioning with Recall Mechanism (61 citations)
- Image Enhanced Event Detection in News Articles (46 citations)
- MULE: Multimodal Universal Language Embedding (37 citations)
and 8 more in the interactive map.
Cross-modal learning · 12 papers
- M3ER: Multiplicative Multimodal Emotion Recognition using Facial, Textual, and Speech Cues (277 citations)
- Modality to Modality Translation: An Adversarial Representation Learning and Graph Fusion Network for Multimodal Fusion (243 citations)
- Urban2Vec: Incorporating Street View Imagery and POIs for Multi-Modal Urban Neighborhood Embedding (116 citations)
- Cross-Modal Attention Network for Temporal Inconsistent Audio-Visual Event Localization (110 citations)
- Hearing Lips: Improving Lip Reading by Distilling Speech Recognizers (97 citations)
- Adaptive Cross-Modal Embeddings for Image-Text Alignment (90 citations)
- Cross-Modal Subspace Clustering via Deep Canonical Correlation Analysis (61 citations)
- Crisis-DIAS: Towards Multimodal Damage Analysis - Deployment, Challenges and Assessment (46 citations)
- Visual Agreement Regularized Training for Multi-Modal Machine Translation (35 citations)
- Semi-Supervised Multi-Modal Learning with Balanced Spectral Decomposition (26 citations)
- Associative Variational Auto-Encoder with Distributed Latent Spaces and Associators (11 citations)
- Learning Multi-Modal Biomarker Representations via Globally Aligned Longitudinal Enrichments (6 citations)
CLIP and zero-shot · 5 papers
- Cross-Lingual Natural Language Generation via Pre-Training (134 citations)
- Attention-Informed Mixed-Language Training for Zero-Shot Cross-Lingual Task-Oriented Dialogue Systems (92 citations)
- Learning Cross-Aligned Latent Embeddings for Zero-Shot Cross-Modal Retrieval (46 citations)
- Mining on Heterogeneous Manifolds for Zero-Shot Cross-Modal Image Retrieval (36 citations)
- A Comparison of Architectures and Pretraining Methods for Contextualized Multilingual Word Embeddings (12 citations)
Multimodal LLMs · 3 papers
- Multimodal Summarization with Guidance of Multimodal Reference (101 citations)
- V-PROM: A Benchmark for Visual Reasoning Using Visual Progressive Matrices (24 citations)
- Region-Based Global Reasoning Networks (3 citations)
