Papers
Topics
Authors
Recent
AI Research Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 86 tok/s
Gemini 2.5 Pro 56 tok/s Pro
GPT-5 Medium 31 tok/s Pro
GPT-5 High 33 tok/s Pro
GPT-4o 102 tok/s Pro
Kimi K2 202 tok/s Pro
GPT OSS 120B 467 tok/s Pro
Claude Sonnet 4 37 tok/s Pro
2000 character limit reached

PaperBridge: Crafting Research Narratives through Human-AI Co-Exploration (2507.14527v1)

Published 19 Jul 2025 in cs.HC

Abstract: Researchers frequently need to synthesize their own publications into coherent narratives that demonstrate their scholarly contributions. To suit diverse communication contexts, exploring alternative ways to organize one's work while maintaining coherence is particularly challenging, especially in interdisciplinary fields like HCI where individual researchers' publications may span diverse domains and methodologies. In this paper, we present PaperBridge, a human-AI co-exploration system informed by a formative study and content analysis. PaperBridge assists researchers in exploring diverse perspectives for organizing their publications into coherent narratives. At its core is a bi-directional analysis engine powered by LLMs, supporting iterative exploration through both top-down user intent (e.g., determining organization structure) and bottom-up refinement on narrative components (e.g., thematic paper groupings). Our user study (N=12) demonstrated PaperBridge's usability and effectiveness in facilitating the exploration of alternative research narratives. Our findings also provided empirical insights into how interactive systems can scaffold academic communication tasks.

Summary

  • The paper presents PaperBridge, an AI-augmented system that synthesizes diverse publications into coherent research narratives.
  • It employs a bi-directional analysis engine combining top-down structured prompts and bottom-up user inputs to generate inspiring narrative perspectives.
  • Evaluation showed excellent usability and enhanced exploration support, highlighting its potential to transform academic narrative construction.

PaperBridge: Crafting Research Narratives through Human-AI Co-Exploration

The paper "PaperBridge: Crafting Research Narratives through Human-AI Co-Exploration" discusses a system called PaperBridge designed to aid researchers, particularly those in Human-Computer Interaction (HCI), in organizing their publications into coherent research narratives. This effort addresses the complex challenge of synthesizing a diverse body of work into narratives that are meaningful across different communication contexts such as job talks, grant proposals, or public engagement scenarios.

System Design and Workflow

PaperBridge integrates human-AI collaboration by providing a bi-directional analysis engine powered by LLMs to support both top-down and bottom-up exploration strategies:

  • Paper Management: The system starts by allowing the user to import their publications and categorize them based on thematic clusters relevant to their research goals.
  • Narrative Exploration: Users can explore different narrative frameworks (e.g., parallel, linear, circular, and coordinate) to see how their work can be interconnected to form various high-level structures.
  • Frameworks and Perspectives: PaperBridge suggests multiple narrative perspectives for each framework, offering contribution statements and thematic clusters to help users frame their research story. The perspectives are provided as "sparks," which are keywords designed to inspire new ways of organizing papers.
  • Organization and Rationale Exploration: Users can refine the system's suggestions by editing themes, re-assigning papers to different clusters, and exploring rationale strategies to justify the importance of their narrative stance. Figure 1

    Figure 1: A narrative perspective is composed of three types of components: a contribution statement, thematic clusters, and individual papers, which are organized within the clusters to articulate the overarching themes.

Bi-directional Analysis Engine

PaperBridge uses a bi-directional analysis engine capable of top-down and bottom-up processes:

  • Top-Down Reasoning: This approach uses structured prompts to instruct LLMs on generating thematic paper clusters. Each framework dictates specific hierarchical relationships that guide the LLM in producing narrative perspectives coherent with the selected framework.
  • Bottom-Up Reasoning: Allows users to start with their own groupings of papers. The system then aligns these groups with existing frameworks, generating new contribution statements and cluster themes accordingly. Figure 2

    Figure 2: Four common narrative frameworks identified from our content analysis. Each framework describes a distinct way of organizing and connecting a researcher's body of work around a central storyline.

Evaluation and Findings

The evaluation included 12 HCI researchers and focused on usability and exploration support. Key findings were:

  • Usability: System Usability Scale (SUS) scores indicated excellent usability, with participants finding the system efficient and intuitive.
  • Exploration Support: Participants appreciated the diverse perspectives and narrative forms generated by PaperBridge, reporting that it helped them explore and understand their research from new angles.
  • Sparks Ratings: Sparks helped by triggering insights even though ratings showed variance based on user familiarity with suggested concepts. Familiar sparks often rated higher, but unfamiliar sparks inspired deeper reflection and alternative narratives. Figure 3

    Figure 3: Assessment of participants' perception of PaperBridge in terms of exploration (Q1-Q4), inspiration (Q5-Q6), and satisfaction (Q7-Q10).

Discussion and Implications

The research highlights the potential of integrating AI with human creativity in academic workflows. The system's design reflects an understanding of the complexity involved in narrative construction for research communication:

  • Frameworks as Cognitive Scaffolds: The defined frameworks serve not only as templates for organizing research but also as cognitive tools that guide users in abstracting and synthesizing their work.
  • Narrative Ideation as Reflective Process: Exploration within PaperBridge sparked introspection, allowing researchers to conceptualize their contributions from fresh perspectives.
  • Challenges: The breadth of papers vis-à-vis thematic coherence affects the generation of sparks, indicating a need for adaptive systems that can tailor suggestions based on the input's diversity.

Conclusion

PaperBridge exemplifies how AI can support the creation of research narratives by enabling structured exploration and reflection on academic work. By facilitating both top-down and bottom-up explorations, it empowers researchers to craft diverse and coherent narratives, adapting their scholarly output to varied audiences and communication needs. Future work can broaden its reach across disciplines, enhancing adaptability, and deepening AI-human interaction models in academic settings.

List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

Youtube Logo Streamline Icon: https://streamlinehq.com

Don't miss out on important new AI/ML research

See which papers are being discussed right now on X, Reddit, and more:

“Emergent Mind helps me see which AI papers have caught fire online.”

Philip

Philip

Creator, AI Explained on YouTube