Efficient Multimodal Generative Recommendation with Latent Narrative Reasoning

arXiv:2609.16070v1 Announce Type: cross Abstract: Generative recommendation reformulates item prediction as semantic identifier generation, yet episodic content introduces a fundamentally different setting where the target is determined by narrative evolution rather than user preference. This task requires models to understand multimodal storyline progression while addressing the efficiency challenges caused by redundant visual contexts and costly explicit reasoning generation. We propose extbf{NarraLite}, an efficient multimodal generative recommendation framework that jointly compresses perception and reasoning. Specifically, Progressive Spectral Compression selectively distills long visual contexts into compact narrative-relevant evidence, preserving transition-critical information while reducing redundant visual computation. Latent Narrative Reasoning introduces context-routed latent reasoning tokens and aligns their contextualized representations with future continuation semantics, enabling implicit narrative inference without autoregressively decoding textual rationales. We further establish a user-agnostic multimodal benchmark for short-form drama continuation across UGC, PGC, and OOD settings. Extensive experiments demonstrate that NarraLite consistently improves continuation accuracy, narrative coherence, and robustness over existing approaches, while achieving a favorable accuracy--efficiency trade-off.

Related

• TopoGR: Revealing and Preserving Latent Structure of Semantic ID in Generative Recommendation
• SynGR: Unleashing the Potential of Cross-Modal Synergy for Generative Recommendation
• WhisperRec: Latent Reasoning for Efficient Foundation Recommendation Models

Source: arXiv cs.AI | 2026-09-16