An official website of the United States government
Writing well requires not only expressing ideas but also refining them through revision, a process facilitated by reflection. Prior research suggests that feedback delivered through dialogues, such as those in writing center tutoring sessions, can help writers reflect more thoughtfully on their work compared to static feedback. Recent advancements in multi-modal large language models (LLMs) now offer new possibilities for supporting interactive and expressive voice-based reflection in writing. In particular, we propose that LLM-generated static feedback can be repurposed as conversation starters, allowing writers to seek clarification, request examples, and ask follow-up questions, thereby fostering deeper reflection on their writing. We argue that voice-based interaction can naturally facilitate this conversational exchange, encouraging writers' engagement with higher-order concerns, facilitating iterative refinement of their reflections, and reduce cognitive load compared to text-based interactions. To investigate these effects, we propose a formative study exploring how text vs. voice input influence writers' reflection and subsequent revisions. Findings from this study will inform the design of intelligent and interactive writing tools, offering insights into how voice-based interactions with LLM-powered conversational agents can support reflection and revision.
more »
« less
This content will become publicly available on May 4, 2026
Interaction-Required Suggestions for Control, Ownership, and Awareness in Human-AI Co-Writing
This paper explores interaction designs for generative AI interfaces that necessitate human involvement throughout the generation process. We argue that such interfaces can promote cognitive engagement, agency, and thoughtful decision-making. Through a case study in text revision, we present and analyze two interaction techniques: (1) using a predictive-text interaction to type the assistant's response to a revision request, and (2) highlighting potential edit opportunities in a document. Our implementations demonstrate how these approaches reveal the landscape of writing possibilities and enable fine-grained control. We discuss implications for human-AI writing partnerships and future interaction design directions.
more »
« less
- Award ID(s):
- 2246145
- PAR ID:
- 10610266
- Publisher / Repository:
- Association for Computational Linguistics
- Date Published:
- Page Range / eLocation ID:
- 62 to 68
- Format(s):
- Medium: X
- Location:
- Albuquerque, New Mexico, US
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Revision plays an important role in writing, and as revisions break down the linearity of the writing process, they are crucial in describing writing process dynamics. Keystroke logging and analysis have been used to identify revisions made during writing. Previous approaches include the manual annotation of revisions, building nonlinear S-notations, and the automated extraction of backspace keypresses. However, these approaches are time-intensive, vulnerable to construct, or restricted. Therefore, this article presents a computational approach to the automatic extraction of full revision events from keystroke logs, including both insertions and deletions, as well as the characters typed to replace the deleted text. Within this approach, revision candidates are first automatically extracted, which allows for a simplified manual annotation of revision events. Second, machine learning is used to automatically detect revision events. For this, 7120 revision events were manually annotated in a dataset of keystrokes obtained from 65 students conducting a writing task. The results showed that revision events could be automatically predicted with a relatively high accuracy. In addition, a case study proved that this approach could be easily applied to a new dataset. To conclude, computational approaches can be beneficial in providing automated insights into revisions in writing.more » « less
-
Abstract As AI systems proliferate, their greenhouse gas emissions are an increasingly important concern for human societies. In this article, we present a comparative analysis of the carbon emissions associated with AI systems (ChatGPT, BLOOM, DALL-E2, Midjourney) and human individuals performing equivalent writing and illustrating tasks. Our findings reveal that AI systems emit between 130 and 1500 times less CO2e per page of text generated compared to human writers, while AI illustration systems emit between 310 and 2900 times less CO2e per image than their human counterparts. Emissions analyses do not account for social impacts such as professional displacement, legality, and rebound effects. In addition, AI is not a substitute for all human tasks. Nevertheless, at present, the use of AI holds the potential to carry out several major activities at much lower emission levels than can humans.more » « less
-
In this paper, Kirchenbauer et. al. use a novel watermarking technology to watermark the output of large language models (LLMs) like ChatGP, which is often in the form of AI-generated text, and mitigate the harms associated with the increasing usage of these technologies. They note some of the capabilities of these LLM models as writing documents, creating executable code, and answering questions, often with human-like capabilities. In addition, they list some of the harms as social engineering and election manipulation campaigns that exploit automated bots on social media platforms, creation of fake news and web content, and use of AI systems for cheating onacademic writing and coding assignments. As for implications for policy makers, this technology can be utilized as a means to regulate and oversee the use of these LLMs on all public and social fronts where their AI-generated text output could pose a potential harm, such as those listed by the authors. (Methods and Metrics, watermarking LLM output)more » « less
-
Abstract Theories and practices to enhance multiple document comprehension and integration are crucial in both personal and work contexts, especially with the proliferation of printed and online sources. This experimental investigation replicates and extends (Wei et al., Educational Technology Research and Development 72:661–685, 2024) to examine how multiple documents integration is influenced by reading purpose, summary writing, and extra-textual networks (pre-reading, Study 1, and post-writing, Study 2). In Study 1 (N = 102), participants were randomly assigned to a pre-reading purpose set by a prompt (integrative or detailed) and by a network (an integrative or else an intra-text network) and then read three documents about Alzheimer’s disease to complete a writing task with revision (but no feedback). Three days later, they completed a delayed writing task and an inference verification test. In Study 2 (N = 90), the same procedure was used except that the network was used as feedback after writing to support revision. Results from the two studies agree with the previous research that the quantity and structural quality of integration can be improved by external cues and by delayed repeated writing. This research further confirms an innovative approach for evaluating different aspects of knowledge integration and contributes to the literature from the concept network perspective as a measure and an intervention of multiple-text reading.more » « less
