# Comparing Sentence-Level Suggestions to Message-Level Suggestions in AI-Mediated Communication

arXiv: https://arxiv.org/abs/2302.13382 · HTML: https://mlsanity.com/paper/2302.13382/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 52 references that are in the map.
- First posted on arXiv 26 Feb 2023.

## Builds on (most cited first, 17 of 17)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Energy and Policy Considerations for Deep Learning in NLP](https://mlsanity.com/paper/1906.02243/) (2019)
- [Scaling Language Models: Methods, Analysis & Insights from Training Gopher](https://mlsanity.com/paper/2112.11446/) (2021)
- [A Diversity-Promoting Objective Function for Neural Conversation Models](https://mlsanity.com/paper/1510.03055/) (2015)
- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [Neural Text Generation with Unlikelihood Training](https://mlsanity.com/paper/1908.04319/) (2019)
- [CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilities](https://mlsanity.com/paper/2201.06796/) (2022)
- [Reducing Sentiment Bias in Language Models via Counterfactual Evaluation](https://mlsanity.com/paper/1911.03064/) (2019)
- [Human heuristics for AI-generated language are flawed](https://mlsanity.com/paper/2206.07271/) (2022)
- [Gmail Smart Compose: Real-Time Assisted Writing](https://mlsanity.com/paper/1906.00080/) (2019)
- [Is GPT-3 Text Indistinguishable from Human Text? Scarecrow: A Framework for Scrutinizing Machine Text](https://mlsanity.com/paper/2107.01294/) (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)
- [Smart Reply: Automated Response Suggestion for Email](https://mlsanity.com/paper/1606.04870/) (2016)
- [Ask Not What AI Can Do, But What AI Should Do: Towards a Framework of Task Delegability](https://mlsanity.com/paper/1902.03245/) (2019)

## Built on by (most cited first, 9 of 9)

- [Beyond the Chat: Executable and Verifiable Text-Editing with LLMs](https://mlsanity.com/paper/2309.15337/) (2023)
- [Understanding the LLM-ification of CHI: Unpacking the Impact of LLMs at CHI through a Systematic Literature Review](https://mlsanity.com/paper/2501.12557/) (2025)
- [Co-Writing with AI, on Human Terms: Aligning Research with User Demands Across the Writing Process](https://mlsanity.com/paper/2504.12488/) (2025)
- [Storyfier: Exploring Vocabulary Learning Support with Text Generation Models](https://mlsanity.com/paper/2308.03864/) (2023)
- [How Does the Disclosure of AI Assistance Affect the Perceptions of Writing?](https://mlsanity.com/paper/2410.04545/) (2024)
- [Understanding and Supporting Formal Email Exchange by Answering AI-Generated Questions](https://mlsanity.com/paper/2502.03804/) (2025)
- [TwIPS: A Large Language Model Powered Texting Application to Simplify Conversational Nuances for Autistic Users](https://mlsanity.com/paper/2407.17760/) (2024)
- [Content-Driven Local Response: Supporting Sentence-Level and Message-Level Mobile Email Replies With and Without AI](https://mlsanity.com/paper/2502.06430/) (2025)
- [Accepted with Minor Revisions: Value of AI-Assisted Scientific Writing](https://mlsanity.com/paper/2511.12529/) (2025)

## Abstract (arXiv)

Traditionally, writing assistance systems have focused on short or even single-word suggestions. Recently, large language models like GPT-3 have made it possible to generate significantly longer natural-sounding suggestions, offering more advanced assistance opportunities. This study explores the trade-offs between sentence- vs. message-level suggestions for AI-mediated communication. We recruited 120 participants to act as staffers from legislators' offices who often need to respond to large volumes of constituent concerns. Participants were asked to reply to emails with different types of assistance. The results show that participants receiving message-level suggestions responded faster and were more satisfied with the experience, as they mainly edited the suggested drafts. In addition, the texts they wrote were evaluated as more helpful by others. In comparison, participants receiving sentence-level assistance retained a higher sense of agency, but took longer for the task as they needed to plan the flow of their responses and decide when to use suggestions. Our findings have implications for designing task-appropriate communication assistance systems.