Demonstrating ViviDoc: Generating Interactive Documents through Human-Agent Collaboration

Yinghao Tang; Yupeng Xie; Yingchaojie Feng; Tingfeng Lan; Wei Chen

cs.CLMar 2, 2026

Demonstrating ViviDoc: Generating Interactive Documents through Human-Agent Collaboration

Authors: Yinghao Tang, Yupeng Xie, Yingchaojie Feng, Tingfeng Lan, Wei Chen

Organizations: State Key Lab of CAD&CG, Zhejiang University · HKUST(GZ) · National University of Singapore · University of Virginia

Abstract

Interactive articles help readers engage with complex ideas through exploration, yet creating them remains costly, requiring both domain expertise and web development skills. Recent LLM-based agents can automate content creation, but naively applying them yields uncontrollable and unverifiable outputs. We present ViviDoc, a human-agent collaborative system that generates interactive educational documents from a single topic input. ViviDoc introduces a multi-agent pipeline (Planner, Executor, Evaluator) and the Document Specification (DocSpec), a human-readable intermediate representation that decomposes each interactive visualization into State, Render, Transition, and Constraint components. The DocSpec enables educators to review and refine generation plans before code is produced, bridging the gap between pedagogical intent and executable output. Expert evaluation and a user study show that ViviDoc substantially outperforms naive agentic generation and provides an intuitive editing experience. Our project homepage is available at https://vividoc-homepage.vercel.app/.

Figures & tables

Figure 1: Three approaches to creating interactive documents: manual authoring, fully automatic generation, and human-agent collaboration through DocSpec.

Figure 2: The ViviDoc pipeline. The Planner generates a Document Specification (DocSpec) from a topic. The user can review and edit the DocSpec before the Executor generates the final document. The Evaluator checks the output and provides feedback.

Figure 3: The ViviDoc user interface. Users follow a Topic $\rightarrow$ Spec $\rightarrow$ Doc workflow to generate interactive educational documents.

Natural language description
“The reader can adjust the radius of a circle using a slider, and as the radius changes, the visualization updates the circle while recalculating and displaying the circumference, diameter, and ratio, allowing the reader to see that the ratio stays roughly the same.”
SRTC Interaction Specification
S	r : slider $[0.5,5]$ , default 1; C : derived $2\pi r$ ; D : derived $2r$ ; ratio : derived $C/D$
R	A circle whose size reflects $r$ ; labels showing $C$ , $D$ , and ratio
T	Dragging the slider changes $r$ ; $C$ , $D$ , ratio update automatically
C	ratio $\approx 3.14159$ regardless of $r$

Table 1: Comparison of a natural language interaction description and its SRTC Interaction Specification for a visualization about

\pi

Figure 4: Expert blind evaluation scores for ViviDoc and Naive Agent across three dimensions.

Figure 5: Participant ratings of ViviDoc on five usability dimensions. Values show mean and standard deviation.

Explore similar work

ViviDoc: Generating Interactive Documents through Human-Agent Collaboration
Mar 30, 2026Yinghao Tang, Yupeng Xie, Yingchaojie Feng +4Human-AI Co-Creation Language Model Generation Evaluation
State Key Lab of CAD&CG, Zhejiang University · HKUST(GZ) · National University of Singapore +2
I-WebGenBench : Evaluating Interactivity in LLM-Generated Scientific Web Applications
May 30, 2026Dasen Dai, Biao Wu, Meng Fang +2AI for Science Web Application Generation
1Vast Intelligence Lab · 2UTS · University of Liverpool
PaperVoyager : Building Interactive Web with Visual Language Models
Mar 24, 2026Dasen Dai, Biao Wu, Meng Fang +1Web Application Generation Document Understanding
Vast Intelligence Lab · UTS · University of Liverpool

Mar 30, 2026J/K move · Enter open · S save

ViviDoc: Generating Interactive Documents through Human-Agent Collaboration

Yinghao Tang, Yupeng Xie, Yingchaojie Feng, Tingfeng Lan+3

State Key Lab of CAD&CG, Zhejiang University · HKUST(GZ) · National University of Singapore+2

Human-AI Co-Creation Language Model Generation Evaluation Multi-Agent LLM Systems LLM Agent Orchestration

Open paper AskPDF arXiv

Interactive documents help readers engage with complex ideas through dynamic visualization, interactive animations, and exploratory interfaces. However, creating such documents remains costly, as it requires both domain expertise and web development skills. Recent Large Language Model (LLM)-based agents can automate content creation, but directly applying them to interactive document generation often produces outputs that are difficult to control. To address this, we present ViviDoc, to the best of our knowledge the first work to systematically address interactive document generation. ViviDoc introduces a multi-agent pipeline (Planner, Styler, Executor, Evaluator). To make the generation process controllable, we provide three levels of human control: (1) the Document Specification (DocSpec) with SRTC Interaction Specifications (State, Render, Transition, Constraint) for structured planning, (2) a content-aware Style Palette for customizing writing and interaction styles, and (3) chat-based editing for iterative refinement. We also construct ViviBench, a benchmark of 101 topics derived from real-world interactive documents across 11 domains, along with a taxonomy of 8 interaction types and a 4-dimensional automated evaluation framework validated against human ratings (Pearson r > 0.84). Experiments show that ViviDoc achieves the highest content richness and interaction quality in both automated and human evaluation. A 12-person user study confirms that the system is easy to use, provides effective control over the generation process, and produces documents that satisfy users.

Abstract

Explore similar work

ViviDoc: Generating Interactive Documents through Human-Agent Collaboration

I-WebGenBench : Evaluating Interactivity in LLM-Generated Scientific Web Applications

PaperVoyager : Building Interactive Web with Visual Language Models