EMNLP Findings2023
Dior-CVAE: Pre-trained Language Models and Diffusion Priors for Variational Dialog Generation
Tianyu Yang, Thy Thy Tran, Iryna Gurevych
Abstract
Current variational dialog models have employed pre-trained language models (PLMs) to parameterize the likelihood and posterior distributions. However, the Gaussian assumption made on the prior distribution is incompatible with these distributions, thus restricting the diversity of generated responses. These models also suffer from posterior collapse, i.e., the decoder tends to ignore latent variables and directly access information captured in the encoder through the cross-attention mechanism.
In this work, we propose Dior-CVAE, a hierarchical conditional variational autoencoder (CVAE) with diffusion priors to address these challenges. We employ a diffusion model to increase the complexity of the prior distribution and its compatibility with the distributions produced by a PLM. Also, we propose memory dropout to the cross-attention mechanism, which actively encourages the use of latent variables for response generation. Our method requires parameters that are comparable to those of previous studies while maintaining comparable inference time, despite the integration of the diffusion model. Overall, experiments across two commonly used open-domain dialog datasets show that our method can generate more diverse responses even without largescale dialog pre-training. Code is available at https://github.com/UKPLab/dior-cvae.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Structured Denoising Diffusion Models in Discrete State-Spaces cites Improved Denoising Diffusion Probabilistic Models
- Structured Denoising Diffusion Models in Discrete State-Spaces cites Denoising Diffusion Probabilistic Models
- Improved Denoising Diffusion Probabilistic Models cites Language Models are Few-Shot Learners
- Improved Denoising Diffusion Probabilistic Models cites Denoising Diffusion Probabilistic Models
- Language Models are Few-Shot Learners cites The Curious Case of Neural Text Degeneration
Builds on28
- Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
- Denoising Diffusion Probabilistic ModelsJonathan Ho, Ajay Jain, Pieter AbbeelNeurIPS 2020 · 35,902 citations
- Improved Denoising Diffusion Probabilistic ModelsAlexander Quinn Nichol, Prafulla DhariwalICML 2021 · 5,234 citations
- The Curious Case of Neural Text DegenerationAri Holtzman, Jan Buys, Li Du, Maxwell Forbes et al.ICLR 2020 · 4,112 citations
- Structured Denoising Diffusion Models in Discrete State-SpacesJacob Austin, Daniel D. Johnson, Jonathan Ho, Daniel Tarlow et al.NeurIPS 2021 · 2,256 citations
Related papers
- DialogVED: A Pre-trained Latent Variable Encoder-Decoder Model for Dialog Response GenerationWei Chen, Yeyun Gong, Song Wang, Bolun Yao et al.ACL 2022
- Controlled Text Generation Using Dictionary Prior in Variational AutoencodersXianghong Fang, Jian Li, Lifeng Shang, Xin Jiang et al.ACL Findings 2022 · 5 citations
- PLATO: Pre-trained Dialogue Generation Model with Discrete Latent VariableSiqi Bao, Huang He, Fan Wang, Hua Wu et al.ACL 2020 · 229 citations
- Towards Diverse, Relevant and Coherent Open-Domain Dialogue Generation via Hybrid Latent VariablesBin Sun, Yitong Li, Fei Mi, Weichao Wang et al.AAAI 2023 · 8 citations
- Generating Relevant and Coherent Dialogue Responses using Self-Separated Conditional Variational AutoEncodersBin Sun, Shaoxiong Feng, Yiwei Li, Jiamou Liu et al.ACL 2021
