# SafeCRS: Personalized Safety Alignment for LLM-Based Conversational Recommender Systems

Source: [SafeCRS: Personalized Safety Alignment for LLM-Based Conversational Recommender Systems](<https://www.sota2.com/research/paper/69a90de92d603e6de3e40679>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-29T05:00:05.450Z

Authors: Haochang Hao, Yifan Xu, Xinzhuo Li, Yingqiang Ge, Lu Cheng

Published: 2026-03-03T21:46:53.000Z

## Abstract

Current LLM-based conversational recommender systems (CRS) primarily optimize recommendation accuracy and user satisfaction. We identify an underexplored vulnerability in which recommendation outputs may negatively impact users by violating personalized safety constraints, when individualized safety sensitivities -- such as trauma triggers, self-harm history, or phobias -- are implicitly inferred from the conversation but not respected during recommendation. We formalize this challenge as personalized CRS safety and introduce SafeRec, a new benchmark dataset designed to systematically evaluate safety risks in LLM-based CRS under user-specific constraints. To further address this problem, we propose SafeCRS, a safety-aware training framework that integrates Safe Supervised Fine-Tuning (Safe-SFT) with Safe Group reward-Decoupled Normalization Policy Optimization (Safe-GDPO) to jointly optimize recommendation quality and personalized safety alignment. Extensive experiments on SafeRec demonstrate that SafeCRS reduces safety violation rates by up to 96.5% relative to the strongest recommendation-quality baseline while maintaining competitive recommendation quality. Warning: This paper contains potentially harmful and offensive content.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Conversational Recommendation on SafeGame](<https://www.sota2.com/research/sota/conversational-recommendation-on-safegame.md>): Conversational Recommendation; SafeGame; Recall@5: 27.02.

- [Conversational Recommendation on SafeMovie](<https://www.sota2.com/research/sota/conversational-recommendation-on-safemovie.md>): Conversational Recommendation; SafeMovie; Recall@5: 7.74.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.03536>)

[PDF](<https://arxiv.org/pdf/2603.03536.pdf>)
