# Beyond Factual Correctness: Mitigating Preference-Inconsistent Explanations in Explainable Recommendation

arXiv: https://arxiv.org/abs/2603.03080 · HTML: https://mlsanity.com/paper/2603.03080/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Preference alignment for recommenders and RAG, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 13 of its 37 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Preference alignment for recommenders and RAG](https://mlsanity.com/alignment/preference-alignment-for-recommenders-and-rag/)

## Builds on (most cited first, 13 of 13)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models](https://mlsanity.com/paper/2405.06211/) (2024)
- [Relational Graph Attention Network for Aspect-based Sentiment Analysis](https://mlsanity.com/paper/2004.12362/) (2020)
- [Personalized Prompt Learning for Explainable Recommendation](https://mlsanity.com/paper/2202.07371/) (2022)
- [Knowledge Graph Retrieval-Augmented Generation for LLM-based Recommendation](https://mlsanity.com/paper/2501.02226/) (2025)
- [Retrieval Augmented Generation with Collaborative Filtering for Personalized Text Generation](https://mlsanity.com/paper/2504.05731/) (2025)
- [KB4Rec: A Dataset for Linking Knowledge Bases with Recommender Systems](https://mlsanity.com/paper/1807.11141/) (2018)
- [Unlocking the Potential of Large Language Models for Explainable Recommendations](https://mlsanity.com/paper/2312.15661/) (2023)
- [Retrieval-Augmented Conversational Recommendation with Prompt-based Semi-Structured Natural Language State Tracking](https://mlsanity.com/paper/2406.00033/) (2024)
- [Quality Metrics in Recommender Systems: Do We Calculate Metrics Consistently?](https://mlsanity.com/paper/2206.12858/) (2022)
- [Neutralizing Popularity Bias in LLM-based Recommendation via Counterfactual Reasoning Guidelines](https://mlsanity.com/paper/2503.08051/) (2025)
- [Consistent Explainers or Unreliable Narrators? Understanding LLM-generated Group Recommendations](https://mlsanity.com/paper/2507.13705/) (2025)
- [Hierarchical Intent-guided Optimization with Pluggable LLM-Driven Semantics for Session-based Recommendation](https://mlsanity.com/paper/2507.04623/) (2025)

## Abstract (arXiv)

LLM-based explainable recommenders can produce fluent explanations that are factually correct, yet still justify items using attributes that conflict with a user's historical preferences. Such preference-inconsistent explanations yield logically valid but unconvincing reasoning and are largely missed by standard hallucination or faithfulness metrics. We formalize this failure mode and propose PURE, a preference-aware reasoning framework following a select-then-generate paradigm. Instead of only improving generation, PURE intervenes in evidence selection, it selects a compact set of multi-hop item-centric reasoning paths that are both factually grounded and aligned with user preference structure, guided by user intent, specificity, and diversity to suppress generic, weakly personalized evidence. The selected evidence is then injected into LLM generation via structure-aware prompting that preserves relational constraints. To measure preference inconsistency, we introduce a feature-level, user-centric evaluation metric that reveals misalignment overlooked by factuality-based measures. Experiments on three real-world datasets show that PURE consistently reduces preference-inconsistent explanations and factual hallucinations while maintaining competitive recommendation accuracy, explanation quality, and inference efficiency. These results highlight that trustworthy explanations require not only factual correctness but also justification aligned with user preferences.