Modeling Text-visual Mutual Dependency for Multi-modal Dialog Generation
arXiv:2105.14445
Abstract
Multi-modal dialog modeling is of growing interest. In this work, we propose frameworks to resolve a specific case of multi-modal dialog generation that better mimics multi-modal dialog generation in the real world, where each dialog turn is associated with the visual context in which it takes place. Specifically, we propose to model the mutual dependency between text-visual features, where the model not only needs to learn the probability of generating the next dialog utterance given preceding dialog utterances and visual contexts, but also the probability of predicting the visual features in which a dialog utterance takes place, leading the generated dialog utterance specific to the visual context. We observe significant performance boosts over vanilla models when the mutual dependency between text and visual features is modeled. Code is available at https://github.com/ShannonAI/OpenViDial.
arXiv commentarXiv admin note: text overlap with arXiv:2012.15015
References in corpus (34)
- Adam: A Method for Stochastic Optimization
- Learning Transferable Visual Models From Natural Language Supervision
- VisualBERT: A Simple and Performant Baseline for Vision and Language
- Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision
- Zero-Shot Text-to-Image Generation
- Deep Reinforcement Learning for Dialogue Generation
- Large-Scale Adversarial Training for Vision-and-Language Representation Learning
- Towards a Human-like Open-Domain Chatbot
- Adversarial Learning for Neural Dialogue Generation
- Unifying Vision-and-Language Tasks via Text Generation
- UNITER: UNiversal Image-TExt Representation Learning
- A Persona-Based Neural Conversation Model
- A Network-based End-to-End Trainable Task-oriented Dialogue System
- Building a Conversational Agent Overnight with Dialogue Self-Play
- A Simple Language Model for Task-Oriented Dialogue
- Image-Grounded Conversations: Multimodal Context for Natural Question and Response Generation
- DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation
- Learning Cooperative Visual Dialog Agents with Deep Reinforcement Learning
- Visual Reference Resolution using Attention Memory for Visual Dialog
- A Neural Network Approach to Context-Sensitive Generation of Conversational Responses
- Adversarial Evaluation of Dialogue Models
- Strategic Dialogue Management via Deep Reinforcement Learning
- DP-GAN: Diversity-Promoting Generative Adversarial Network for Generating Informative and Diversified Text
- M6: A Chinese Multimodal Pretrainer
- Alternating Recurrent Dialog Model with Large-scale Pre-trained Language Models
- Multi-step Reasoning via Recurrent Dual Attention for Visual Dialog
- XGPT: Cross-modal Generative Pre-Training for Image Captioning
- OpenViDial: A Large-Scale, Open-Domain Dialogue Dataset with Visual Contexts
- Factor Graph Attention
- Dual Attention Networks for Visual Reference Resolution in Visual Dialog
- A Tailored Pre-Training Model for Task-Oriented Dialog Generation
- Zero-shot Generalization in Dialog State Tracking through Generative Question Answering
- Generating More Interesting Responses in Neural Conversation Models with Distributional Constraints
- Teaching Machines to Converse