Research map / Label-efficient and robust learning / Recognition and long-tail
Deep Interest Evolution Network for Click-Through Rate Prediction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 DIEN's interest extraction and target-aware interest evolution to the two-layer GRU attention baseline | Two layer GRU Attention · Electronics | AUC: 0.7605±0.00059 -> 0.7792±0.00243 | Helped |
| Add DIEN's interest extraction and target-aware interest evolution to the two-layer GRU attention baseline | Two layer GRU Attention · Books | AUC: 0.7890±0.00268 -> 0.8453±0.00476 | Helped |
| Add DIEN's interest extraction and target-aware interest evolution to the two-layer GRU attention baseline | Two layer GRU Attention · Industrial dataset | AUC: 0.6457 -> 0.6541 | Helped |
| Replace AIGRU with AUGRU for interest evolution (ablation) | BaseModel + GRU interest model · Electronics | AUC: 0.7606±0.00061 -> 0.7640±0.00073 | Helped |
| Replace AIGRU with AUGRU for interest evolution (ablation) | BaseModel + GRU interest model · Books | AUC: 0.7892±0.00222 -> 0.7911±0.00150 | Helped |
| Replace AGRU with AUGRU for interest evolution (ablation) | BaseModel + GRU interest model · Electronics | AUC: 0.7628±0.00015 -> 0.7640±0.00073 | Helped |
| Replace AGRU with AUGRU for interest evolution (ablation) | BaseModel + GRU interest model · Books | AUC: 0.7890±0.00268 -> 0.7911±0.00150 | Helped |
| Add the auxiliary loss to the GRU+AUGRU model (ablation) | BaseModel + GRU + AUGRU · Electronics | AUC: 0.7640±0.00073 -> 0.7792±0.00243 | Helped |
| Add the auxiliary loss to the GRU+AUGRU model (ablation) | BaseModel + GRU + AUGRU · Books | AUC: 0.7911±0.00150 -> 0.8453±0.00476 | Helped |
| Add the auxiliary loss to the GRU+AUGRU model (ablation) | BaseModel + GRU + AUGRU · Industrial dataset | AUC: 0.6493 -> 0.6541 | Helped |
| Deploy DIEN instead of BaseModel | Taobao display advertising system · Online A/B testing | CTR: + 20.7% | Helped |
| Deploy DIEN instead of BaseModel | Taobao display advertising system · Online A/B testing | PPC: - 3.0% | Hurt |
Closest papers by meaning
- From Feature Interaction to Feature Generation: A Generative Paradigm of CTR Prediction Models (ICML 2025 · similarity 0.85)
- Addressing Cold-Start Problem in Click-Through Rate Prediction via Supervised Diffusion Modeling (AAAI 2025 · similarity 0.85)
- AT4CTR: Auxiliary Match Tasks for Enhancing Click-Through Rate Prediction (AAAI 2024 · similarity 0.84)
- Generalize for Future: Slow and Fast Trajectory Learning for CTR Prediction (AAAI 2024 · similarity 0.84)
- UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence (NeurIPS 2026 · similarity 0.83)
- Listwise Preference Diffusion Optimization for User Behavior Trajectories Prediction (NeurIPS 2025 · similarity 0.82)
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.