# Safety as a Constraint: Fine-Tuning a LLM Recommender to Explain Itself

arXiv: https://arxiv.org/abs/2609.13657 · HTML: https://mlsanity.com/paper/2609.13657/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 39 references that are in the map.
- First posted on arXiv 12 Sep 2026.

## Builds on (most cited first, 28 of 28)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [Prometheus: Inducing Fine-grained Evaluation Capability in Language Models](https://mlsanity.com/paper/2310.08491/) (2023)
- [Reward Constrained Policy Optimization](https://mlsanity.com/paper/1805.11074/) (2018)
- [Interpretable Preferences via Multi-Objective Reward Modeling and Mixture-of-Experts](https://mlsanity.com/paper/2406.12845/) (2024)
- [JudgeLM: Fine-tuned Large Language Models are Scalable Judges](https://mlsanity.com/paper/2310.17631/) (2023)
- [Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards](https://mlsanity.com/paper/2306.04488/) (2023)
- [Explainable Recommendation: A Survey and New Perspectives](https://mlsanity.com/paper/1804.11192/) (2018)
- [MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoning](https://mlsanity.com/paper/2310.16049/) (2023)
- [Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation](https://mlsanity.com/paper/2311.09049/) (2023)
- [Measuring Association Between Labels and Free-Text Rationales](https://mlsanity.com/paper/2010.12762/) (2020)
- [GuardReasoner: Towards Reasoning-based LLM Safeguards](https://mlsanity.com/paper/2501.18492/) (2025)
- [Confronting Reward Model Overoptimization with Constrained RLHF](https://mlsanity.com/paper/2310.04373/) (2023)
- [Personalized Transformer for Explainable Recommendation](https://mlsanity.com/paper/2105.11601/) (2021)
- [XRec: Large Language Models for Explainable Recommendation](https://mlsanity.com/paper/2406.02377/) (2024)
- [Controllable and Diverse Text Generation in E-commerce](https://mlsanity.com/paper/2102.11497/) (2021)
- [Contextualizing Recommendation Explanations with LLMs: A User Study](https://mlsanity.com/paper/2501.12152/) (2025)

## Abstract (arXiv)

Traditional recommender systems are typically trained to predict what item users will interact with next, but not why. However, offering personalized evidence for why a user might like the predicted item is an important way to enhance the service and to raise the likelihood that the user will be genuinely interested in the recommendation. This service can be delivered by integrating a frontier-model call into the member-facing pipeline, but it will add extra cost and latency. In this paper, we train a recommender LLM to generate personalized explanations for its reccomendation, based on the user's watching history at a large video streaming service. We impose two requirements on the generated explanation: it must be faithful to the elements of the shows it links, and it must be strictly non-harmful to the user. To this end, we first train two LLM-judge reward models covering three specific criteria, and propose constrained GRPO to incorporate these different criteria. On a held-out real-world testing set, our fine-tuned model improves the all-three-criteria PASS rate rises from 0.649 to 0.956 under our own judges and from 0.677 to 0.931 under an independent judge, where as the frontier generator performs similar to the untuned recommender baseline. We conduct further experiments to show that the model's language and recommendation abilities remain unchanged. Based on these results, we conclude that an LLM-based recommender can be fine-tuned on other complex tasks without compromising its original recommendation performance, thus provide insights for further agentic user interface powered by a single model.