Research map / Label-efficient and robust learning / Recognition and long-tail
DFRec: Dual Fluctuation Modeling of Multi-level Intent Evolution for Next-Item Recommendation
Nengjun Zhu, Lingdan Sun, Qi Zhang, Jian Cao, Hang Yu
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 |
|---|---|---|---|
| add Emerging Intent Generation Module (EIGM) (ablation) | DFRec · Yoochoose, Jdata, Diginetica | Recall and NDCG | Improved |
| add long-term fluctuation variance in DIFM (ablation) | DFRec · Yoochoose, Jdata, Diginetica | Recall and NDCG | Improved |
| add short-term fluctuation variance in DIFM (ablation) | DFRec · Yoochoose, Jdata, Diginetica | Recall and NDCG | Improved |
| fuse multi-level continuous item segments instead of single items (ablation) | DFRec · Yoochoose, Jdata, Diginetica | Recall and NDCG | Improved |
| use window size Nw 2 or 3 instead of small or large | DFRec · Yoochoose and Jdata | Recall and NDCG | Mixed |
| use DFRec instead of MiaSRec | MiaSRec · Jdata | Recall@5: 43.53 -> 44.60 | Improved |
| use DFRec instead of MiaSRec | MiaSRec · Yoochoose | NDCG@5: 67.64 -> 69.48 | Improved |
| use DFRec instead of MiaSRec | MiaSRec · Diginetica | Recall@10: 79.31 -> 79.40 | Improved |
| use intent-based models instead of non-intent SR-GNN | SR-GNN · Jdata | NDCG@5: 27.81 -> 32.09 | Improved |
| train DFRec for more epochs (diversity) | DFRec · Yoochoose and Jdata | Diversity | Improved |
Closest papers by meaning
- Multi-granularity Intent Modeling with Adversarial Robustness for Sequential Recommendation (AAAI 2026 · similarity 0.89)
- Intent Oriented Contrastive Learning for Sequential Recommendation (AAAI 2025 · similarity 0.87)
- De-collapsing User Intent: Adaptive Diffusion Augmentation with Mixture-of-Experts for Sequential Recommendation (AAAI 2026 · similarity 0.87)
- Ready for You When You Are Back: Content-Driven Session-Based Recommendation for Continuity of Experience (AAAI 2025 · similarity 0.85)
- Evidential Stochastic Differential Equations for Time-Aware Sequential Recommendation (NeurIPS 2024 · similarity 0.85)
- On the Unexpected Effectiveness of Reinforcement Learning for Sequential Recommendation (ICML 2024 · 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.
