RELEASE · MODELS · #702
DSRec: dual-interest cross-SSM for sequential recommendation
arXiv:2609.21548v1 introduces DSRec, a dual-interest cross-SSM architecture that disentangles long-term and short-term item roles by encoding items into long-term interest embeddings (processed by a full-sequence Mamba SSM) and a time-modulated short-term SSM, with a residual cross-fusion mechanism to align the two branches; experiments on public benchmarks report improvements over prior state-of-the-art methods.
KEY POINTS
- arXiv:2609.21548v1 introduces DSRec, a dual-interest cross-SSM architecture that disentangles long-term and short-term item roles by encoding items into long-term interest embeddings (processed by a full-sequence Mamba SSM) and a time-modulated short-term SSM, with a residual cross-fusion mechanism to align the two branches; experiments on public benchmarks report improvements over prior state-of-the-art methods.
- By modeling item polysemy and aligning long- and short-term interests with efficient SSMs, DSRec targets improved and more cost-effective long-range sequential recommendation.
- Dual-Interest Sequential Product Recommendation With Multi-Granular SSM
WHY IT MATTERS
By modeling item polysemy and aligning long- and short-term interests with efficient SSMs, DSRec targets improved and more cost-effective long-range sequential recommendation.