MoS applies theme-aware routing to extract multi-scale theme-specific subsequences from noisy long user sequences, achieving state-of-the-art recommendation performance with fewer FLOPs than comparable MoE models.
arXiv preprint arXiv:2505.14728 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Mixture of Sequence: Theme-Aware Mixture-of-Experts for Long-Sequence Recommendation
MoS applies theme-aware routing to extract multi-scale theme-specific subsequences from noisy long user sequences, achieving state-of-the-art recommendation performance with fewer FLOPs than comparable MoE models.
- NormAct: Benchmarking Embodied Agents' Proactive Compliance with Unspoken Social Norms