Towards Trustworthy LLM-Based Recommendation via Rationale Integration

TL;DR

This paper introduces LLM-Rec, a recommendation system that generates logical rationales before suggesting items, improving transparency and accuracy by leveraging a rationale-first approach and chain-of-thought reasoning, validated on Amazon datasets.

Contribution

The paper presents a novel LLM-based recommender that generates grounded rationales prior to recommendations, enhancing interpretability and performance over existing methods.

Findings

Significant performance improvements on Amazon Fashion and Scientific domains.

Enhanced interpretability through rationale generation.

Public release of a rationale-augmented recommendation dataset.

Abstract

Traditional recommender systems (RS) have been primarily optimized for accuracy and short-term engagement, often overlooking transparency and trustworthiness. Recently, platforms such as Amazon and Instagram have begun providing recommendation rationales to users, acknowledging their critical role in fostering trust and enhancing engagement; however, most existing systems still treat them as post-hoc artifacts. We propose an LLM-based recommender (LLM-Rec) that not only predicts items but also generates logically grounded rationales. Our approach leverages a self-annotated rationale dataset and instruction tuning in a rationale-first format, where the model generates an explanation before outputting the recommended item. By adopting this strategy and representing rationales in a chain-of-thought (CoT) style, LLM-Rec strengthens both interpretability and recommendation performance. Experiments on the Fashion and Scientific domains of the Amazon Review dataset demonstrate significant improvements over well-established baselines. To encourage reproducibility and future research, we publicly release a rationale-augmented recommendation dataset containing user histories, rationales, and recommended items.