Research map / Label-efficient and robust learning / Recognition and long-tail
Attention-over-Attention Field-Aware Factorization Machine
Zhibo Wang, Jinxin Ma, Yongquan Zhang, Qian Wang, Ju Ren, Peng Sun
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 AFM attention with attention-over-attention (ablation) | FM embeddings (AFM) · Movielens validation | RMSE: 0.4339 -> 0.4248 | Improved |
| replace AFM attention with attention-over-attention (ablation) | FM embeddings (AFM) · Frappe validation | RMSE: 0.3089 -> 0.3068 | Improved |
| replace AFM attention with attention-over-attention (ablation) | FFM embeddings (AFFM) · Movielens validation | RMSE: 0.4178 -> 0.4128 | Improved |
| replace AFM attention with attention-over-attention (ablation) | FFM embeddings (AFFM) · Frappe validation | RMSE: 0.3130 -> 0.3116 | Improved |
| use field-aware embeddings (FFM) instead of FM embeddings under AoA attention (ablation) | AoAFM · Movielens test | RMSE: 0.4234 -> 0.4126 | Improved |
| use field-aware embeddings (FFM) instead of FM embeddings under AoA attention (ablation) | AoAFM · Frappe test | RMSE: 0.3112 -> 0.3141 | Hurt |
| AoAFFM vs FFM (add AoA attention) | FFM · Criteo test | RMSE: 0.3851 -> 0.3808 | Improved |
| AoAFFM vs FFM (add AoA attention) | FFM · Movielens test | RMSE: 0.4478 -> 0.4126 | Improved |
| AoAFFM vs FFM (add AoA attention) | FFM · Frappe test | RMSE: 0.3345 -> 0.3141 | Improved |
| increase regularization strength from 1 to 10 (ablation) | AoAFFM/AFM · Movielens validation | RMSE | Hurt |
| add dropout at proper ratio (ablation) | AFM, AoAFFM, FM, FFM · Movielens/Frappe validation | RMSE | Improved |
Closest papers by meaning
- Interaction-Aware Factorization Machines for Recommender Systems (AAAI 2019 · similarity 0.90)
- fastFM: A Library for Factorization Machines (JMLR 2016 · similarity 0.86)
- Boosting Factorization Machines via Saliency-Guided Mixup (TPAMI 2024 · similarity 0.85)
- Tensorized Attention for Understanding Multi-Object Relationships (AAAI 2025 · similarity 0.82)
- Sparse Attention as Compact Kernel Regression (NeurIPS 2026 · similarity 0.82)
- From Feature Interaction to Feature Generation: A Generative Paradigm of CTR Prediction Models (ICML 2025 · similarity 0.81)
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.
