PeReGrINE: Evaluating Personalized Review Fidelity with User Item Graph Context
This work addresses the need for reproducible evaluation of personalized review generation for researchers in natural language processing and recommendation systems, but it is incremental as it builds on existing retrieval-conditioned language models and datasets.
The authors tackled the problem of evaluating personalized review generation by introducing PeReGrINE, a benchmark and framework that restructures Amazon Reviews 2023 into a graph for controlled evidence comparison, and found that graph-derived evidence primarily drives personalization and consistency, with visual evidence sometimes improving textual quality.
We introduce PeReGrINE, a benchmark and evaluation framework for personalized review generation grounded in graph-structured user--item evidence. PeReGrINE restructures Amazon Reviews 2023 into a temporally consistent bipartite graph, where each target review is conditioned on bounded evidence from user history, item context, and neighborhood interactions under explicit temporal cutoffs. To represent persistent user preferences without conditioning directly on sparse raw histories, we compute a User Style Parameter that summarizes each user's linguistic and affective tendencies over prior reviews. This setup supports controlled comparison of four graph-derived retrieval settings: product-only, user-only, neighbor-only, and combined evidence. Beyond standard generation metrics, we introduce Dissonance Analysis, a macro-level evaluation framework that measures deviation from expected user style and product-level consensus. We also study visual evidence as an auxiliary context source and find that it can improve textual quality in some settings, while graph-derived evidence remains the main driver of personalization and consistency. Across product categories, PeReGrINE offers a reproducible way to study how evidence composition affects review fidelity, personalization, and grounding in retrieval-conditioned language models.