# Gmail Smart Compose: Real-Time Assisted Writing

arXiv: https://arxiv.org/abs/1906.00080 · HTML: https://mlsanity.com/paper/1906.00080/

## Key facts

- Cited by 98 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 May 2019.

## Built on by (most cited first, 30 of 98)

- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [Membership Inference Attacks From First Principles](https://mlsanity.com/paper/2112.03570/) (2021)
- [Analyzing Leakage of Personally Identifiable Information in Language Models](https://mlsanity.com/paper/2302.00539/) (2023)
- [Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs](https://mlsanity.com/paper/2210.13312/) (2022)
- [Adversary Instantiation: Lower Bounds for Differentially Private Machine Learning](https://mlsanity.com/paper/2101.04535/) (2021)
- [What Does it Mean for a Language Model to Preserve Privacy?](https://mlsanity.com/paper/2202.05520/) (2022)
- [Large Pre-trained Language Models Contain Human-like Biases of What is Right and Wrong to Do](https://mlsanity.com/paper/2103.11790/) (2021)
- [Co-Writing with Opinionated Language Models Affects Users' Views](https://mlsanity.com/paper/2302.00560/) (2023)
- [SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents](https://mlsanity.com/paper/2403.08715/) (2024)
- [Evaluating Human-Language Model Interaction](https://mlsanity.com/paper/2212.09746/) (2022)
- [Pearl: Personalizing Large Language Model Writing Assistants with Generation-Calibrated Retrievers](https://mlsanity.com/paper/2311.09180/) (2023)
- [A Study of Face Obfuscation in ImageNet](https://mlsanity.com/paper/2103.06191/) (2021)
- [Sparks: Inspiration for Science Writing using Language Models](https://mlsanity.com/paper/2110.07640/) (2021)
- [Follow My Instruction and Spill the Beans: Scalable Data Extraction from Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2402.17840/) (2024)
- [A Design Space for Intelligent and Interactive Writing Assistants](https://mlsanity.com/paper/2403.14117/) (2024)
- [Towards Facilitating Empathic Conversations in Online Mental Health Support: A Reinforcement Learning Approach](https://mlsanity.com/paper/2101.07714/) (2021)
- [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)
- [AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuances](https://mlsanity.com/paper/2409.11360/) (2024)
- [Shaping Human-AI Collaboration: Varied Scaffolding Levels in Co-writing with Language Models](https://mlsanity.com/paper/2402.11723/) (2024)
- [Beyond Text Generation: Supporting Writers with Continuous Automatic Text Summaries](https://mlsanity.com/paper/2208.09323/) (2022)
- [Zero-Shot Recommender Systems](https://mlsanity.com/paper/2105.08318/) (2021)
- [Memorization in NLP Fine-tuning Methods](https://mlsanity.com/paper/2205.12506/) (2022)
- [Quantum federated learning through blind quantum computing](https://mlsanity.com/paper/2103.08403/) (2021)
- [FRUIT: Faithfully Reflecting Updated Information in Text](https://mlsanity.com/paper/2112.08634/) (2021)
- [Choice Over Control: How Users Write with Large Language Models using Diegetic and Non-Diegetic Prompting](https://mlsanity.com/paper/2303.03199/) (2023)
- [User Inference Attacks on Large Language Models](https://mlsanity.com/paper/2310.09266/) (2023)
- [Membership Inference on Word Embedding and Beyond](https://mlsanity.com/paper/2106.11384/) (2021)
- [Privacy Implications of Retrieval-Based Language Models](https://mlsanity.com/paper/2305.14888/) (2023)
- [Training and Evaluating a Jupyter Notebook Data Science Assistant](https://mlsanity.com/paper/2201.12901/) (2022)
- [Privacy Regularization: Joint Privacy-Utility Optimization in Language Models](https://mlsanity.com/paper/2103.07567/) (2021)

## Abstract (arXiv)

In this paper, we present Smart Compose, a novel system for generating interactive, real-time suggestions in Gmail that assists users in writing mails by reducing repetitive typing. In the design and deployment of such a large-scale and complicated system, we faced several challenges including model selection, performance evaluation, serving and other practical issues. At the core of Smart Compose is a large-scale neural language model. We leveraged state-of-the-art machine learning techniques for language model training which enabled high-quality suggestion prediction, and constructed novel serving infrastructure for high-throughput and real-time inference. Experimental results show the effectiveness of our proposed system design and deployment approach. This system is currently being served in Gmail.