TY - GEN
T1 - IdeaSynth
T2 - 2025 CHI Conference on Human Factors in Computing Systems, CHI 2025
AU - Pu, Kevin
AU - Feng, K. J.Kevin
AU - Grossman, Tovi
AU - Hope, Tom
AU - Dalvi Mishra, Bhavana
AU - Latzke, Matt
AU - Bragg, Jonathan
AU - Chang, Joseph Chee
AU - Siangliulue, Pao
N1 - Publisher Copyright: © 2025 Copyright held by the owner/author(s).
PY - 2025/4/26
Y1 - 2025/4/26
N2 - Research ideation involves broad exploring and deep refining ideas. Both require deep engagement with literature. Existing tools focus primarily on broad idea generation, yet offer little support for iterative specification, refinement, and evaluation needed to further develop initial ideas. To bridge this gap, we introduce IdeaSynth, a research idea development system that uses LLMs to provide literature-grounded feedback for articulating research problems, solutions, evaluations, and contributions. IdeaSynth represents these idea facets as nodes on a canvas, and allow researchers to iteratively refine them by creating and exploring variations and combinations. Our lab study (N = 20) showed that participants, while using IdeaSynth, explored more alternative ideas and expanded initial ideas with more details compared to a strong LLM-based baseline. Our deployment study (N = 7) demonstrated that participants effectively used IdeaSynth for real-world research projects at various ideation stages from developing initial ideas to revising framings of mature manuscripts, highlighting the possibilities to adopt IdeaSynth in researcher's workflows.
AB - Research ideation involves broad exploring and deep refining ideas. Both require deep engagement with literature. Existing tools focus primarily on broad idea generation, yet offer little support for iterative specification, refinement, and evaluation needed to further develop initial ideas. To bridge this gap, we introduce IdeaSynth, a research idea development system that uses LLMs to provide literature-grounded feedback for articulating research problems, solutions, evaluations, and contributions. IdeaSynth represents these idea facets as nodes on a canvas, and allow researchers to iteratively refine them by creating and exploring variations and combinations. Our lab study (N = 20) showed that participants, while using IdeaSynth, explored more alternative ideas and expanded initial ideas with more details compared to a strong LLM-based baseline. Our deployment study (N = 7) demonstrated that participants effectively used IdeaSynth for real-world research projects at various ideation stages from developing initial ideas to revising framings of mature manuscripts, highlighting the possibilities to adopt IdeaSynth in researcher's workflows.
KW - Human-AI Collaboration
KW - Research Ideation
KW - Scientific Literature
UR - http://www.scopus.com/inward/record.url?scp=105005729883&partnerID=8YFLogxK
U2 - 10.1145/3706598.3714057
DO - 10.1145/3706598.3714057
M3 - منشور من مؤتمر
T3 - Conference on Human Factors in Computing Systems - Proceedings
BT - CHI 2025 - Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems
Y2 - 26 April 2025 through 1 May 2025
ER -