Steering AI-Driven Personalization of Scientific Text for General Audiences

2024-11-15

arXiv.org

AI Summary

To address significant disparities in audience expertise, cultural background, and cognitive preferences in science communication, this study proposes a dynamic, user-profile–driven personalization method—leveraging education level, interests, and geography—and introduces TranSlider, an interactive tool enabling real-time adjustment of scientific text “affinity” via a 0–100 slider to jointly optimize information density and comprehensibility. Key contributions include: (1) the first multi-gradient, continuously adjustable translation framework; (2) empirical identification of cumulative gains in comprehension and scientific trust as affinity increases incrementally; and (3) a dual-path adaptation model balancing concision and contextualization. An LLM-powered user study (n=15) demonstrates that high-affinity versions significantly enhance readability and emotional resonance, low-affinity versions preserve domain-specific precision, and concurrent exposure to multiple affinity variants deepens holistic understanding and trust.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Machine Learning: Learning Preferences or Rankings
• Knowledge Representation and Reasoning: Preferences

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Digital media platforms (e.g., social media, science blogs) offer opportunities to communicate scientific content to general audiences at scale. However, these audiences vary in their scientific expertise, literacy levels, and personal backgrounds, making effective science communication challenging. To address this challenge, we designed TranSlider, an AI-powered tool that generates personalized translations of scientific text based on individual user profiles (e.g., hobbies, location, and education). Our tool features an interactive slider that allows users to steer the degree of personalization from 0 (weakly relatable) to 100 (strongly relatable), leveraging LLMs to generate the translations with given degrees. Through an exploratory study with 15 participants, we investigated both the utility of these AI-personalized translations and how interactive reading features influenced users' understanding and reading experiences. We found that participants who preferred higher degrees of personalization appreciated the relatable and contextual translations, while those who preferred lower degrees valued concise translations with subtle contextualization. Furthermore, participants reported the compounding effect of multiple translations on their understanding of scientific content. Given these findings, we discuss several implications of AI-personalized translation tools in facilitating communication in collaborative contexts.

Problem

• Personalizing scientific text for diverse general audiences
• Balancing relatability and conciseness in AI translations
• Enhancing science communication via steerable human-AI interfaces

Innovation

• AI tool generates personalized scientific text translations
• Interactive slider controls personalization degree (0-100)
• LLMs create translations based on user profiles