Apple has revealed how it’s using generative AI to create review summaries on the App Store, a feature introduced with iOS 18.4. Designed to give users a quick overview of what others think about an app, these summaries are crafted using a multi-step, large language model (LLM) system fine-tuned with LoRA adapters.
“To generate accurate and useful summaries of reviews in the App Store, our system addresses a number of challenges, including the dynamic nature of this multi-document environment and the diversity of user reviews,” said Apple in a post on its machine learning portal.
The AI filters spam and irrelevant content before extracting key insights from real user reviews. It then identifies major themes, prioritizes app-related feedback, and generates a brief, balanced summary. Apple says it fine-tunes the summaries to be safe, accurate, helpful, and consistent with its voice.
To keep the summaries relevant, the system dynamically updates as new reviews roll in, while ensuring diversity and grounding in actual user sentiment. Human raters evaluate thousands of samples for quality across safety, groundedness, composition, and helpfulness.
Apple emphasised that the goal isn’t just to speed up browsing but to help users make better download decisions, while showing how AI can responsibly handle summarizing vast, messy user-generated content. “In addition to delivering useful summaries for App Store users, this work more broadly demonstrates the potential of LLM-based summarization to enhance decision-making in high-volume, user-generated content settings,” said Apple.
Discover the latest Business News, Sensex, and Nifty updates. Obtain Personal Finance insights, tax queries, and expert opinions on Moneycontrol or download the Moneycontrol App to stay updated!
