# Diversifying Reply Suggestions using a Matching-Conditional Variational Autoencoder

arXiv: https://arxiv.org/abs/1903.10630 · HTML: https://mlsanity.com/paper/1903.10630/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Mar 2019.

## Built on by (most cited first, 14 of 14)

- [The Impact of Multiple Parallel Phrase Suggestions on Email Input and Composition Behaviour of Native and Non-Native English Writers](https://mlsanity.com/paper/2101.09157/) (2021)
- [Fast and Memory Efficient Differentially Private-SGD via JL Projections](https://mlsanity.com/paper/2102.03013/) (2021)
- ["One-Size-Fits-All"? Examining Expectations around What Constitute "Fair" or "Good" NLG System Behaviors](https://mlsanity.com/paper/2310.15398/) (2023)
- [A Dataset and Baselines for Multilingual Reply Suggestion](https://mlsanity.com/paper/2106.02017/) (2021)
- [Differentially Private n-gram Extraction](https://mlsanity.com/paper/2108.02831/) (2021)
- [Combing for Credentials: Active Pattern Extraction from Smart Reply](https://mlsanity.com/paper/2207.10802/) (2022)
- [Who Wrote this? How Smart Replies Impact Language and Agency in the Workplace](https://mlsanity.com/paper/2210.06470/) (2022)
- [Language Scaling for Universal Suggested Replies Model](https://mlsanity.com/paper/2106.02232/) (2021)
- [A Conditional Generative Matching Model for Multi-lingual Reply Suggestion](https://mlsanity.com/paper/2109.07046/) (2021)
- [Exploring Low-Cost Transformer Model Compression for Large-Scale Commercial Reply Suggestions](https://mlsanity.com/paper/2111.13999/) (2021)
- [Few-Max: Few-Shot Domain Adaptation for Unsupervised Contrastive Representation Learning](https://mlsanity.com/paper/2206.10137/) (2022)
- [Model-Based Simulation for Optimising Smart Reply](https://mlsanity.com/paper/2305.16852/) (2023)
- [Enhancing AI Assisted Writing with One-Shot Implicit Negative Feedback](https://mlsanity.com/paper/2410.11009/) (2024)
- [Mixture of Decoders for Diverse Dialog Response Generation](https://mlsanity.com/paper/2610.00621/) (2026)

## Abstract (arXiv)

We consider the problem of diversifying automated reply suggestions for a commercial instant-messaging (IM) system (Skype). Our conversation model is a standard matching based information retrieval architecture, which consists of two parallel encoders to project messages and replies into a common feature representation. During inference, we select replies from a fixed response set using nearest neighbors in the feature space. To diversify responses, we formulate the model as a generative latent variable model with Conditional Variational Auto-Encoder (M-CVAE). We propose a constrained-sampling approach to make the variational inference in M-CVAE efficient for our production system. In offline experiments, M-CVAE consistently increased diversity by ~30-40% without significant impact on relevance. This translated to a 5% gain in click-rate in our online production system.