arXiv:2507.00783v2 Announce Type: replace Abstract: In this paper, we contribute to the debate on generative artificial intelligence (GenAI) in scientometrics. We argue that moving from a trial-and-error approach to an explainable and actionable use requires a principled understanding of strengths and weaknesses of GenAI as compared with other techniques and with human judgment. To this end, we introduce a conceptual framework based on the distinction between the semantic dimensions of texts, i.e. the meanings attributed to words, and their pragmatic dimension, i.e. their embedding within comm
The proliferation of generative AI tools necessitates a more rigorous and principled approach to their application in academic and professional fields like scientometrics.
This paper highlights the growing need for explainability and a deep understanding of generative AI's strengths and weaknesses, moving beyond superficial application to informed integration.
The focus shifts from simply experimenting with generative AI to developing conceptual frameworks that distinguish its semantic and pragmatic dimensions, influencing how its impacts are assessed.
- · AI ethicists
- · Academics in scientometrics
- · Organizations developing explainable AI
- · Researchers embracing principled AI use
- · Ad-hoc AI tool developers
- · Organizations using GenAI without critical understanding
- · Trial-and-error AI integration strategies
The academic discourse around generative AI applications becomes more sophisticated and nuanced.
Development of new metrics and methodologies for evaluating GenAI's output in specialized fields, leading to more robust and reliable AI-driven insights.
Increased trust and adoption of carefully vetted generative AI systems, potentially accelerating their deep integration across various professional domains where accuracy and explainability are paramount.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL