EviSnap: Faithful Evidence-Cited Explanations for Cold-Start Cross-Domain Recommendation

TL;DR

EviSnap is a novel cross-domain recommendation framework that provides faithful, evidence-based explanations for its predictions by leveraging compact, interpretable concept representations derived from reviews.

Contribution

It introduces a lightweight, interpretable CDR model that generates evidence-cited rationales and enables exact score decomposition and counterfactual explanations.

Findings

EviSnap outperforms baseline models on Amazon Reviews transfers.

It produces explanations that pass deletion and sufficiency faithfulness tests.

The model enables counterfactual 'what-if' scenario analysis.

Abstract

Cold-start cross-domain recommender (CDR) systems predict a user's preferences in a target domain using only their source-domain behavior, yet existing CDR models either map opaque embeddings or rely on post-hoc or LLM-generated rationales that are hard to audit. We introduce EviSnap a lightweight CDR framework whose predictions are explained by construction with evidence-cited, faithful rationales. EviSnap distills noisy reviews into compact facet cards using an LLM offline, pairing each facet with verbatim supporting sentences. It then induces a shared, domain-agnostic concept bank by clustering facet embeddings and computes user-positive, user-negative, and item-presence concept activations via evidence-weighted pooling. A single linear concept-to-concept map transfers users across domains, and a linear scoring head yields per-concept additive contributions, enabling exact score decompositions and counterfactual 'what-if' edits grounded in the cited sentences. Experiments on the Amazon Reviews dataset across six transfers among Books, Movies, and Music show that EviSnap consistently outperforms strong mapping and review-text baselines while passing deletion- and sufficiency-based tests for explanation faithfulness.