Information Studies

Permanent URI for this communityhttp://hdl.handle.net/1903/2249

Browse

Author: search.filters.author.Srinivasan, Arvind

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Fluid Transformers and Creative Analogies: Exploring Large Language Models’ Capacity for Augmenting Cross-Domain Analogical Creativity
    (Association for Computer Machinery (ACM), 2023-06-19) Ding, Zijian; Srinivasan, Arvind; MacNeil, Stephen; Chan, Joel
    Cross-domain analogical reasoning is a core creative ability that can be challenging for humans. Recent work has shown some proofsof-concept of Large language Models’ (LLMs) ability to generate cross-domain analogies. However, the reliability and potential usefulness of this capacity for augmenting human creative work has received little systematic exploration. In this paper, we systematically explore LLMs capacity to augment cross-domain analogical reasoning. Across three studies, we found: 1) LLM-generated crossdomain analogies were frequently judged as helpful in the context of a problem reformulation task (median 4 out of 5 helpfulness rating), and frequently (∼80% of cases) led to observable changes in problem formulations, and 2) there was an upper bound of ∼25% of outputs being rated as potentially harmful, with a majority due to potentially upsetting content, rather than biased or toxic content. These results demonstrate the potential utility — and risks — of LLMs for augmenting cross-domain analogical creativity.
  • Thumbnail Image
    Item
    Improving Selection of Analogical Inspirations with Chunking and Recombination
    (2023) Srinivasan, Arvind; Chan, Joel; Library & Information Services; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Innovation is vital in various fields, and analogical thinking is a powerful tool for gen- erating creative solutions to complex problems. However, recognizing analogies can be time- consuming, and successful recognition doesn’t guarantee their adoption in innovation. In this thesis, A novel computational support system for analogical innovation is proposed that employs the cognitive mechanisms for chunking and recombination as mediums of interaction. Chunking involves identifying and extracting meaningful chunks or segments from a design problem into interactive tiles called magnets while recombination involves combining these magnets to gener- ate insightful questions that elicit divergent thinking. In this way, the proposed system aims to streamline the process of recognizing and selecting analogical inspirations for innovation while avoiding premature rejection and design fixation.To evaluate the effectiveness of the system, a within-subjects study involving 23 participants was conducted, comparing the proposed interface with a baseline. The study found that using chunking and recombination as interactive mechanisms helped prevent premature rejection of useful analogical leads, resulting in 4 times fewer ignored analogical leads. Participants were also found to make 12 times fewer changes to their decisions, given a minor increment in processing time in the order of 1.5 minutes. Overall, these results suggest that our proposed intervention is an effective tool for facilitating the selection of beneficial analogies, fostering analogical innovation through computational support.