Research map / Architectures and efficiency / Attention and transformers
Sequential Recommendation with Relation-Aware Kernelized Self-Attention
Mingi Ji, Weonyoung Joo, Kyungwoo Song, Yoon-Yeong Kim, Il-Chul Moon
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace deterministic self-attention with RKSA's stochastic, relation-aware kernelized attention | SASRec-style sequential recommendation · Beauty | Hit@10: 0.4720 -> 0.5015 | Improved |
| Replace deterministic self-attention with RKSA's stochastic, relation-aware kernelized attention | SASRec-style sequential recommendation · Games | Hit@10: 0.7373 -> 0.7551 | Improved |
| Replace deterministic self-attention with RKSA's stochastic, relation-aware kernelized attention | SASRec-style sequential recommendation · CiteULike | Hit@10: 0.6757 -> 0.6893 | Improved |
| Replace deterministic self-attention with RKSA's stochastic, relation-aware kernelized attention | SASRec-style sequential recommendation · Steam | Hit@10: 0.8610 -> 0.8668 | Improved |
| Replace deterministic self-attention with RKSA's stochastic, relation-aware kernelized attention | SASRec-style sequential recommendation · MovieLens | Hit@10: 0.5873 -> 0.5997 | Improved |
| Use the counting kernel rather than item/user representation kernels (ablation) | RKSA · Beauty | Hit@10 | Improved |
| Use the item and user kernels rather than relying on counting-kernel information (ablation) | RKSA · MovieLens | Hit@10 | Improved |
| Emphasize counting, item, or user relations according to the dataset's properties (ablation) | RKSA kernel combination · Beauty and MovieLens | Hit@10 | Mixed |
Closest papers by meaning
- An Attentive Inductive Bias for Sequential Recommendation beyond the Self-Attention (AAAI 2024 · similarity 0.89)
- Iterative Sparse Attention for Long-sequence Recommendation (AAAI 2025 · similarity 0.87)
- TV-Rec: Time-Variant Convolutional Filter for Sequential Recommendation (NeurIPS 2025 · similarity 0.87)
- LLM4RSR: Large Language Models as Data Correctors for Robust Sequential Recommendation (AAAI 2025 · similarity 0.85)
- LIME: Link-based User-item Interaction Modeling with Decoupled XOR Attention for Efficient Test Time Scaling (NeurIPS 2026 · similarity 0.85)
- SLMRec: Distilling Large Language Models into Small for Sequential Recommendation (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
