Rethinking Semantic Collaborative Integration: Why Alignment Is Not Enough

TL;DR

This paper challenges the common assumption that semantic and collaborative representations in recommender systems should be globally aligned, proposing instead a view that emphasizes their complementarity and heterogeneity.

Contribution

It introduces a shared-plus-private latent structure model and complementarity-aware diagnostics, advocating for a shift from alignment to fusion-based integration in LLM-enhanced recommenders.

Findings

Low item-level agreement between semantic and collaborative views

Significant oracle fusion gains indicate strong complementarity

Low-capacity mappings only capture shared components, especially under distribution shift

Abstract

Large language models (LLMs) have become an important semantic infrastructure for modern recommender systems. A prevailing paradigm integrates LLM-derived semantic embeddings with collaborative representations via representation alignment, implicitly assuming that the two views encode a shared latent entity and that stronger alignment yields better results. We formalize this assumption as the global low-complexity alignment hypothesis and argue that it is stronger than necessary and often structurally mismatched with real-world recommendation settings. We propose a complementary perspective in which semantic and collaborative representations are treated as partially shared yet fundamentally heterogeneous views, each containing both shared and view-specific factors. Under this shared-plus-private latent structure, enforcing global geometric alignment may distort local structure, suppress view-specific signals, and reduce informational diversity. To support this perspective, we develop complementarity-aware diagnostics that quantify overlap, unique-hit contribution, and theoretical fusion upper bounds. Empirical analyses on sparse recommendation benchmarks reveal low item-level agreement between semantic and collaborative views and substantial oracle fusion gains, indicating strong complementarity. Furthermore, controlled alignment probes show that low-capacity mappings capture only shared components and fail to recover full collaborative geometry, especially under distribution shift. These findings suggest that alignment should not be treated as the default integration principle. We advocate a shift from alignment-centric modeling to complementarity fusion-centric, complementarity-aware design, where shared factors are selectively integrated while private signals are preserved. This reframing provides a principled foundation for the next generation of LLM-enhanced recommender systems.