# Towards Diverse, Relevant and Coherent Open-Domain Dialogue Generation via Hybrid Latent Variables

arXiv: https://arxiv.org/abs/2212.01145 · HTML: https://mlsanity.com/paper/2212.01145/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 34 references that are in the map.
- First posted on arXiv 2 Dec 2022.

## Builds on (most cited first, 28 of 28)

- [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)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Sequence to Sequence Learning with Neural Networks](https://mlsanity.com/paper/1409.3215/) (2014)
- [Unified Language Model Pre-training for Natural Language Understanding and Generation](https://mlsanity.com/paper/1905.03197/) (2019)
- [DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset](https://mlsanity.com/paper/1710.03957/) (2017)
- [Generating Sentences from a Continuous Space](https://mlsanity.com/paper/1511.06349/) (2015)
- [Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders](https://mlsanity.com/paper/1703.10960/) (2017)
- [PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable](https://mlsanity.com/paper/1910.07931/) (2019)
- [Neural Responding Machine for Short-Text Conversation](https://mlsanity.com/paper/1503.02364/) (2015)
- [Adversarial Learning for Neural Dialogue Generation](https://mlsanity.com/paper/1701.06547/) (2017)
- [A Neural Network Approach to Context-Sensitive Generation of Conversational Responses](https://mlsanity.com/paper/1506.06714/) (2015)
- [Attribute2Image: Conditional Image Generation from Visual Attributes](https://mlsanity.com/paper/1512.00570/) (2015)
- [PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning](https://mlsanity.com/paper/2006.16779/) (2020)
- [DialogWAE: Multimodal Response Generation with Conditional Wasserstein Auto-Encoder](https://mlsanity.com/paper/1805.12352/) (2018)
- [Transformer-based Conditional Variational Autoencoder for Controllable Story Generation](https://mlsanity.com/paper/2101.00828/) (2021)
- [Variational Transformers for Diverse Response Generation](https://mlsanity.com/paper/2003.12738/) (2020)
- [DialogVED: A Pre-trained Latent Variable Encoder-Decoder Model for Dialog Response Generation](https://mlsanity.com/paper/2204.13031/) (2022)
- [Diverse and Accurate Image Description Using a Variational Auto-Encoder with an Additive Gaussian Encoding Space](https://mlsanity.com/paper/1711.07068/) (2017)
- [Improving Neural Conversational Models with Entropy-Based Data Filtering](https://mlsanity.com/paper/1905.05471/) (2019)
- [Jointly Optimizing Diversity and Relevance in Neural Response Generation](https://mlsanity.com/paper/1902.11205/) (2019)
- [A Batch Normalized Inference Network Keeps the KL Vanishing Away](https://mlsanity.com/paper/2004.12585/) (2020)
- [Fuse It More Deeply! A Variational Transformer with Layer-Wise Latent Variable Inference for Text Generation](https://mlsanity.com/paper/2207.06130/) (2022)
- [Generating Relevant and Coherent Dialogue Responses using Self-separated Conditional Variational AutoEncoders](https://mlsanity.com/paper/2106.03410/) (2021)
- [A Discrete CVAE for Response Generation on Short-Text Conversation](https://mlsanity.com/paper/1911.09845/) (2019)
- [A Conditional Variational Framework for Dialog Generation](https://mlsanity.com/paper/1705.00316/) (2017)
- [PanGu-Bot: Efficient Generative Dialogue Pre-training from Pre-trained Language Model](https://mlsanity.com/paper/2203.17090/) (2022)
- [Modeling Complex Dialogue Mappings via Sentence Semantic Segmentation Guided Conditional Variational Auto-Encoder](https://mlsanity.com/paper/2212.00231/) (2022)

## Built on by (most cited first, 5 of 5)

- [Focused Large Language Models are Stable Many-Shot Learners](https://mlsanity.com/paper/2408.13987/) (2024)
- [Enhancing User-Oriented Proactivity in Open-Domain Dialogues with Critic Guidance](https://mlsanity.com/paper/2505.12334/) (2025)
- [What Would You Ask When You First Saw a2+b2=c2? Evaluating LLM on Curiosity-Driven Questioning](https://mlsanity.com/paper/2409.17172/) (2024)
- [Dior-CVAE: Pre-trained Language Models and Diffusion Priors for Variational Dialog Generation](https://mlsanity.com/paper/2305.15025/) (2023)
- [Multi-Dimensional Prompt Chaining to Improve Open-Domain Dialogue Generation](https://mlsanity.com/paper/2601.01037/) (2026)

## Abstract (arXiv)

Conditional variational models, using either continuous or discrete latent variables, are powerful for open-domain dialogue response generation. However, previous works show that continuous latent variables tend to reduce the coherence of generated responses. In this paper, we also found that discrete latent variables have difficulty capturing more diverse expressions. To tackle these problems, we combine the merits of both continuous and discrete latent variables and propose a Hybrid Latent Variable (HLV) method. Specifically, HLV constrains the global semantics of responses through discrete latent variables and enriches responses with continuous latent variables. Thus, we diversify the generated responses while maintaining relevance and coherence. In addition, we propose Conditional Hybrid Variational Transformer (CHVT) to construct and to utilize HLV with transformers for dialogue generation. Through fine-grained symbolic-level semantic information and additive Gaussian mixing, we construct the distribution of continuous variables, prompting the generation of diverse expressions. Meanwhile, to maintain the relevance and coherence, the discrete latent variable is optimized by self-separation training. Experimental results on two dialogue generation datasets (DailyDialog and Opensubtitles) show that CHVT is superior to traditional transformer-based variational mechanism w.r.t. diversity, relevance and coherence metrics. Moreover, we also demonstrate the benefit of applying HLV to fine-tuning two pre-trained dialogue models (PLATO and BART-base).