# The Dialog Must Go On: Improving Visual Dialog via Generative Self-Training

arXiv: https://arxiv.org/abs/2205.12502 · HTML: https://mlsanity.com/paper/2205.12502/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 85 of its 102 references that are in the map.
- First posted on arXiv 25 May 2022.

## Builds on (most cited first, 30 of 85)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks](https://mlsanity.com/paper/1908.02265/) (2019)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015)
- [MixMatch: A Holistic Approach to Semi-Supervised Learning](https://mlsanity.com/paper/1905.02249/) (2019)
- [Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering](https://mlsanity.com/paper/1707.07998/) (2017)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts](https://mlsanity.com/paper/2102.08981/) (2021)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [Unsupervised Data Augmentation for Consistency Training](https://mlsanity.com/paper/1904.12848/) (2019)
- [Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books](https://mlsanity.com/paper/1506.06724/) (2015)
- [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://mlsanity.com/paper/1911.00536/) (2019)
- [Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment](https://mlsanity.com/paper/1907.11932/) (2019)
- [DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset](https://mlsanity.com/paper/1710.03957/) (2017)
- [BERT-ATTACK: Adversarial Attack Against BERT Using BERT](https://mlsanity.com/paper/2004.09984/) (2020)
- [Visual Dialog](https://mlsanity.com/paper/1611.08669/) (2016)
- [A Deep Reinforced Model for Abstractive Summarization](https://mlsanity.com/paper/1705.04304/) (2017)
- [Rethinking Pre-training and Self-training](https://mlsanity.com/paper/2006.06882/) (2020)

## Built on by (most cited first, 7 of 7)

- [InfoVisDial: An Informative Visual Dialogue Dataset by Bridging Large Multimodal and Language Models](https://mlsanity.com/paper/2312.13503/) (2023)
- [mathbbVD-mathbbGR: Boosting mathbbVisual mathbbDialog with Cascaded Spatial-Temporal Multi-Modal mathbbGRaphs](https://mlsanity.com/paper/2310.16590/) (2023)
- [PROGrasp: Pragmatic Human-Robot Communication for Object Grasping](https://mlsanity.com/paper/2309.07759/) (2023)
- [Direct Diffusion Score Preference Optimization via Stepwise Contrastive Policy-Pair Supervision](https://mlsanity.com/paper/2512.23426/) (2025)
- [Context-Aware Multi-Turn Visual-Textual Reasoning in LVLMs via Dynamic Memory and Adaptive Visual Guidance](https://mlsanity.com/paper/2509.05669/) (2025)
- [From text to multimodal: a survey of adversarial example generation in question answering systems](https://mlsanity.com/paper/2312.16156/) (2023)
- [Synthesizing Sentiment-Controlled Feedback For Multimodal Text and Image Data](https://mlsanity.com/paper/2402.07640/) (2024)

## Abstract (arXiv)

Visual dialog (VisDial) is a task of answering a sequence of questions grounded in an image, using the dialog history as context. Prior work has trained the dialog agents solely on VisDial data via supervised learning or leveraged pre-training on related vision-and-language datasets. This paper presents a semi-supervised learning approach for visually-grounded dialog, called Generative Self-Training (GST), to leverage unlabeled images on the Web. Specifically, GST first retrieves in-domain images through out-of-distribution detection and generates synthetic dialogs regarding the images via multimodal conditional text generation. GST then trains a dialog agent on the synthetic and the original VisDial data. As a result, GST scales the amount of training data up to an order of magnitude that of VisDial (1.2M to 12.9M QA data). For robust training of the synthetic dialogs, we also propose perplexity-based data selection and multimodal consistency regularization. Evaluation on VisDial v1.0 and v0.9 datasets shows that GST achieves new state-of-the-art results on both datasets. We further observe the robustness of GST against both visual and textual adversarial attacks. Finally, GST yields strong performance gains in the low-data regime. Code is available at https://github.com/gicheonkang/gst-visdial.