LLM-Assisted Writing and Authoring

会議の名前
CHI 2026
Cerebra: Aligning Implicit Knowledge in Interactive SQL Authoring
要旨

LLM-driven tools have significantly lowered barriers to writing SQL queries. However, user instructions are often underspecified, assuming the model understands implicit knowledge, such as dataset schemas, domain conventions, and task-specific requirements, that isn't explicitly provided. This results in frequently erroneous scripts that require users to repeatedly clarify their intent. Additionally, users struggle to validate generated scripts because they cannot verify whether the model correctly applied implicit knowledge. We present Cerebra, an interactive NL-to-SQL tool that aligns implicit knowledge between users and LLMs during SQL authoring. Cerebra automatically retrieves implicit knowledge from historical SQL scripts based on user instructions, presents this knowledge in an interactive tree view for code review, and supports iterative refinement to improve generated scripts. To evaluate the effectiveness and usability of Cerebra, we conducted a user study with 16 participants, demonstrating its improved support for customized SQL authoring. The source code of Cerebra is available at https://github.com/zjuidg/CHI26-Cerebra.

著者
Yunfan Zhou
Zhejiang University, Hangzhou, Zhejiang, China
Qiming Shi
Zhejiang University, Hangzhou, Zhejiang, China
Zhongsu Luo
Zhejiang University, Hangzhou, Zhejiang, China
Xiwen Cai
China Mobile, Shenzhen, China
Yanwei Huang
HKUST, Hong Kong S.A.R., China
Dae Hyun Kim
Yonsei University, Seoul, Korea, Republic of
Di Weng
Zhejiang University, Ningbo, Zhejiang, China
Yingcai Wu
Zhejiang University, Hangzhou, Zhejiang, China
Beyond the Spotlight: Co-Designing AI for Theatre Audience Communication
要旨

Theatres and concert halls play a crucial role within the performing arts, where managerial and administrative staff are essential to bringing live performances to audiences. Existing AI research has focused on artistic creation, but less attention has been paid to the purposeful design of AI systems that support organisational practices. This paper addresses this gap by identifying the needs, challenges and opportunities for AI integration into everyday workflows, forming the basis for design principles to guide the architecturing, training, and deployment of AI systems that empower staff, rather than replace them. This is explored through a co-design workshop with theatre marketing and communication professionals. Through reflections of the themes explored in the workshop and by following the guiding principles, this paper presents examples of implementation of AI systems that could be adopted, offering concrete directions for developing AI that benefits the cultural sector.

著者
Alan Pedrassoli Chitayat
University of York, York, United Kingdom
Steph Carter
University of York, York, United Kingdom
Barry Alan. Robertson
The Grey Hill, Loans, United Kingdom
Jonathan Hook
University of York, York, United Kingdom
iRULER: Intelligible Rubric-Based User-Defined LLM Evaluation for Revision
要旨

Large Language Models (LLMs) have become indispensable for evaluating writing. However, text feedback they provide is often unintelligible, generic, and not specific to user criteria. Inspired by structured rubrics in education and intelligible AI explanations, we propose iRULER following identified design guidelines to \textit{scaffold} the review process by \textit{specific} criteria, providing \textit{justification} for score selection, and offering \textit{actionable} revisions to target different quality levels. To \textit{qualify} user-defined criteria, we recursively used iRULER with a rubric-of-rubrics to iteratively \textit{refine} rubrics. In controlled experiments on writing revision and rubric creation, iRULER most improved validated LLM-judged review scores and was perceived as most helpful and aligned compared to read-only rubric and text-based LLM feedback. Qualitative findings further support how iRULER satisfies the design guidelines for user-defined feedback. This work contributes interactive rubric tools for intelligible LLM-based review and revision of writing, and user-defined rubric creation.

著者
Jingwen Bai
National University of Singapore, Singapore, Singapore
Wei Soon Cheong
National University of Singapore, Singapore, Singapore
Philippe Muller
IRIT, Toulouse, France
Brian Y. Lim
National University of Singapore, Singapore, Singapore
動画
Narrix: Remixing Narrative Strategies from Examples for Story Writing
要旨

Experienced storytellers decompose stories into local narrative strategies and how these strategies shape higher-level arcs. This decomposition helps writers recognize patterns in others' work and adapt those patterns to tell new stories. Novices, however, struggle to identify these strategies or to reuse them effectively. We present Narrix, a novel writing tool that helps novice writers recognize narrative strategies in example stories and repurpose these strategies in their own writing. Narrix analyzes strategies in example stories, highlights them with color-coded lexical cues and explanations, and situates them on an interactive story arc for exploration by emotional shifts and turning points. Writers then drag strategies onto multi-dimensional tracks and apply block-scoped edits to revise or continue their drafts through controlled generation steered by specified strategies. Through a within-subjects study (N=12), Narrix showed improved participants' retention, confidence, and creative adaptation of narrative strategies compared to a baseline chat-based writing interface.

著者
Chao Zhang
Cornell University, Ithaca, New York, United States
Shunan Guo
Adobe Research, San Jose, California, United States
Abe Davis
Cornell University, New York, New York, United States
Eunyee Koh
Adobe Research, San Jose, California, United States
DiaryPlay: AI-Assisted Creation of Interactive Story Vignettes for Everyday Storytelling
要旨

An interactive vignette is a visual storytelling medium that lets the audience role-play a character and interact with non-player characters (NPCs) and the digital environment. Yet, the authoring complexity of interactive vignettes has obstructed their adoption in everyday storytelling, which builds on immediacy. We introduce DiaryPlay, an AI-assisted authoring system that generates interactive vignettes from text stories. The Authoring Component visually elicits three core elements (environment, characters, events) through automation and author refinement. The Viewing Component delivers an interactive story to the audience using an LLM-powered Controlled Divergence Module, which allows divergent player and NPC behaviors within the boundaries defined by the author's intended story. A technical evaluation shows that the Controlled Divergence module generates believable NPC activities based on both character persona and storyline. A user study demonstrates that DiaryPlay enables low-effort authoring of interactive vignettes for everyday storytelling while providing engaging viewing experiences and conveying the core story message.

著者
Jiangnan Xu
Tampere University, Tampere, Finland
Haeseul Cha
KAIST, Daejeon, Korea, Republic of
Gosu Choi
KAIST, Daejeon, Korea, Republic of
Gyu-cheol Lee
Korea Telecom, Seoul, Korea, Republic of
Yeo-Jin Yoon
Korea Telecom, Seoul, Korea, Republic of
Zucheul Lee
Korea Telecom, Seoul, Korea, Republic of
Konstantinos Papangelis
Rochester Institute of Technology, Rochester, New York, United States
Dae Hyun Kim
Yonsei University, Seoul, Korea, Republic of
Juho Kim
KAIST, Daejeon, Korea, Republic of
Linting Style and Substance in READMEs
要旨

READMEs shape first impressions of software projects, yet what constitutes a good README varies across audiences and contexts. Research software needs reproducibility details, while open-source libraries might prioritize quick-start guides. Through a design probe, LintMe, we explore how linting can be used to improve READMEs given these diverse contexts, aiding style and content issues while preserving authorial agency. Users create context-specific checks using a lightweight DSL that uses a novel combination of programmatic operations (e.g., for broken links) with LLM-based content evaluation (e.g., for detecting jargon), yielding checks that would be challenging for prior linters. Through a user study (N=11), comparison with naive LLM usage, and an extensibility case study, we find that our design is approachable, flexible, and well matched with the needs of this domain. This work opens the door for linting more complex documentation and other culturally mediated text-based documents.

著者
Hima Mynampaty
University of Utah, Salt Lake City, Utah, United States
Nathania Josephine
University of Utah, Salt Lake City, Utah, United States
Katherine E.. Isaacs
The University of Utah, Salt Lake City, Utah, United States
Andrew M. McNutt
University of Utah, Salt Lake City, Utah, United States
動画
Orchid-Creator: An Authoring Tool Supporting LLM-Driven Interactive Narrative Creation
要旨

Large language models (LLMs) are reshaping interactive digital narratives (IDNs). However, creating complex interactive narratives while preserving narrative consistency remains challenging. We present Orchid‑Creator (Orchid), an LLM-based authoring tool that represents IDNs as story graphs with a card‑based interface for scene definition and conditional transitions. We evaluated Orchid in two studies: a usability study with eight authors, and a comparative study with 20 participants (authors, developers, and players) that compared Orchid to Twine and AI Dungeon. Authors reported that Orchid’s features met their needs (card‑based interface: 4.0/5; story graph: 4.38/5; variable setup: 4.5/5). Structuring narratives with Orchid was easier (M = 6.0/7, p < .01) and produced better‑structured stories (M = 5.3/7, p < .05) than the alternatives, balancing author control (M = 5.5/7) with outcome diversity (M = 5/7, p < .01) and maintaining comparable usability. Finally, a case study with an artist demonstrates Orchid’s utility for interactive art.

著者
Zhen Wu
HKUST, HongKong, China
serkan kumyol
Hong Kong University of Science and Technology, Hong Kong, China
Zhengyang Ma
Hong Kong University of Science and Technology, Hong Kong, Hong Kong
Tristan Braud
The Hong Kong University of Science and Technology, ClearWater Bay, Kowloon, Hong Kong