Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
December 5, 2025ACM Transactions on Interactive Intelligent Systems

User Perceptions of Personalized and Generic Explanations in LLM-Driven Recommender Systems

View Full Paper
Ask AI
Bookmark
Share

Authors

ÍSÍtallo De Sousa SilvaLBLeandro BalbyASAlan Said

Discussion

Loading...

Member takes

Overview

User study assesses transparency and interpretability in ChatGPT-generated recommendations, suggesting personalization impacts effectiveness in unfamiliar items.

Key Points

  • Participants rated ChatGPT's recommendations higher than random ones, supporting its effectiveness in generating user-centered explanations.
  • When unfamiliar with a recommended movie, users found personalized explanations more impactful compared to generic ones, highlighting the context of personalization.
  • The study included 94 participants evaluating the transparency and interpretability of different types of recommendations using ChatGPT.
  • Findings suggest that while LLMs can personalize explanations, their effectiveness can vary based on users' prior knowledge of the items.

Cite This Study

Silva et al. (2025) studied this question.

synapsesocial.com/papers/693231368e51979591dcebe7https://doi.org/10.1145/3779059
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Leveraging ChatGPT for Automated Human-centered Explanations in Recommender Systems2024 · 35 citations
  2. 2Large Language Models as Evaluators for Recommendation Explanations2024
  3. 3LLM-Powered Explanations: Unraveling Recommendations Through Subgraph Reasoning2024 · 3 citations
  4. 4Navigating User Experience of ChatGPT-based Conversational Recommender Systems: The Effects of Prompt Guidance and Recommendation Domain2024 · 3 citations
  5. 5Consistent Explainers or Unreliable Narrators: Systematic Differences in Consistency and Sensitivity Across Large Language Models for Group Recommendations2026