Transparent and Scrutable Recommendations Using Natural Language User Profiles
Jerome Ramos, Hossen A. Rahmani, Xi Wang, Xiao Fu, Aldo Lipani
TL;DR
The paper tackles the opacity of contemporary recommender systems by replacing opaque user embeddings with natural-language user profiles generated from past reviews via instruction-tuned LLM prompts. It trains a scrutable NL-based recommender that predicts ratings using only NL profiles, enabling direct scrutiny and edits to steer recommendations. Across two benchmark datasets, the NL-profile approach achieves competitive warm-start performance while providing transparent, user-editable control over preferences, with a demonstrated ability to adjust outputs by simple NL-profile edits. The work highlights practical benefits for user trust and autonomy and outlines future directions for multi-turn updates and bias safeguards in transparent recommendation systems.
Abstract
Recent state-of-the-art recommender systems predominantly rely on either implicit or explicit feedback from users to suggest new items. While effective in recommending novel options, many recommender systems often use uninterpretable embeddings to represent user preferences. This lack of transparency not only limits user understanding of why certain items are suggested but also reduces the user's ability to scrutinize and modify their preferences, thereby affecting their ability to receive a list of preferred recommendations. Given the recent advances in Large Language Models (LLMs), we investigate how a properly crafted prompt can be used to summarize a user's preferences from past reviews and recommend items based only on language-based preferences. In particular, we study how LLMs can be prompted to generate a natural language (NL) user profile that holistically describe a user's preferences. These NL profiles can then be leveraged to fine-tune a LLM using only NL profiles to make transparent and scrutable recommendations. Furthermore, we validate the scrutability of our user profile-based recommender by investigating the impact on recommendation changes after editing NL user profiles. According to our evaluations of the model's rating prediction performance on two benchmarking rating prediction datasets, we observe that this novel approach maintains a performance level on par with established recommender systems in a warm-start setting. With a systematic analysis into the effect of updating user profiles and system prompts, we show the advantage of our approach in easier adjustment of user preferences and a greater autonomy over users' received recommendations.