SOURCE-LINKED INTELLIGENCE

Latent-Aligned Reasoning for Multimodal Recommendation

arXiv · AI, language, vision and robotics · article · Sep 4, 2026 · UTC

Multimodal Vision-Language Models (VLMs) have demonstrated remarkable capabilities in cross-modal understanding, yet a fundamental challenge persists when applying them to recommendation: as representations propagate through multi-step reasoning, both visual and textual signals progressively attenuate - a phenomenon we term cross-modal dilution. To address this, we propose LARK (Latent-Aligned Reasoning frameworK), a two-stage latent reasoning framework with complementary alignment mechanisms within a single VLM. In the first stage, learnable latent tokens are interleaved with multi-step chain

Read original source ↗ Open in workspace

recordType

paper

region

Global