# Transparent and Scrutable Recommendations Using Natural Language User Profiles

arXiv: https://arxiv.org/abs/2402.05810 · HTML: https://mlsanity.com/paper/2402.05810/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Feb 2024.

## Built on by (most cited first, 24 of 24)

- [Value Profiles for Encoding Human Variation](https://mlsanity.com/paper/2503.15484/) (2025)
- [End-to-end Training for Recommendation with Language-based User Profiles](https://mlsanity.com/paper/2410.18870/) (2024)
- [ProAgent: Harnessing On-Demand Sensory Contexts for Proactive LLM Agent Systems in the Wild](https://mlsanity.com/paper/2512.06721/) (2025)
- [POPI: Personalizing LLMs via Optimized Natural Language Preference Inference](https://mlsanity.com/paper/2510.17881/) (2025)
- [TEARS: Textual Representations for Scrutable Recommendations](https://mlsanity.com/paper/2410.19302/) (2024)
- [PREF: Reference-Free Evaluation of Personalised Text Generation in LLMs](https://mlsanity.com/paper/2508.10028/) (2025)
- [CTRL-Rec: Controlling Recommender Systems With Natural Language](https://mlsanity.com/paper/2510.12742/) (2025)
- [GISTBench: Evaluating LLM User Understanding via Evidence-Based Interest Verification](https://mlsanity.com/paper/2603.29112/) (2026)
- [When Ads Become Profiles: Uncovering the Invisible Risk of Web Advertising at Scale with LLMs](https://mlsanity.com/paper/2509.18874/) (2025)
- [Enriching Semantic Profiles into Knowledge Graph for Recommender Systems Using Large Language Models](https://mlsanity.com/paper/2601.08148/) (2026)
- [SensorPersona: An LLM-Empowered System for Continual Persona Extraction from Longitudinal Mobile Sensor Streams](https://mlsanity.com/paper/2604.06204/) (2026)
- [DeMeVa at LeWiDi-2025: Modeling Perspectives with In-Context Learning and Label Distribution Learning](https://mlsanity.com/paper/2509.09524/) (2025)
- [SteerEval: A Framework for Evaluating Steerability with Natural Language Profiles for Recommendation](https://mlsanity.com/paper/2601.21105/) (2026)
- [Efficient Personalization of Generative User Interfaces](https://mlsanity.com/paper/2604.09876/) (2026)
- [Hierarchical Multi-Persona Induction from User Behavioral Logs: Learning Evidence-Grounded and Truthful Personas](https://mlsanity.com/paper/2604.26120/) (2026)
- [Bridging Textual Profiles and Latent User Embeddings for Personalization](https://mlsanity.com/paper/2605.06981/) (2026)
- [PaperFlow: Profiling, Recommending, and Adapting Across Daily Paper Streams](https://mlsanity.com/paper/2606.07454/) (2026)
- [Paying to Know: Micro-Transaction Markets for Verified Product Information in Agentic E-Commerce](https://mlsanity.com/paper/2606.24783/) (2026)
- [Inverse Theory of Mind Modeling for Content Recommendation: From Web Browsing to Dynamic Intelligent Interfaces](https://mlsanity.com/paper/2608.11354/) (2026)
- [Dear Algo: A Precision-First Agentic Intent Layer for Unified Search and Recommendation](https://mlsanity.com/paper/2608.15877/) (2026)
- [Profiling What Matters: Context-Aware Item Profiles from Large-Scale Metadata for LLM Recommenders](https://mlsanity.com/paper/2608.20801/) (2026)
- [Inferring Hidden User Models from the Behavior of Personalized LLM Agents](https://mlsanity.com/paper/2609.03815/) (2026)
- [Reproducing Transparent and Scrutable Recommendations: Exploring Open-Weight Models via Natural-Language User Profiles](https://mlsanity.com/paper/2609.19831/) (2026)
- [From "Who Is This User?" to "What Does This Purchase Mean?": A Deployed Pipeline for Semantic User Profiling at Bank Scale](https://mlsanity.com/paper/2609.19928/) (2026)

## Abstract (arXiv)

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.