Research map / Architectures and efficiency / RNNs and normalization
Leveraging Title-Abstract Attentive Semantics for Paper Recommendation
Guibing Guo, Bowei Chen, Xiaoyan Zhang, Zhirong Liu, Zhenhua Dong, Xiuqiang He
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 word-level title-aware attention (vs average of LSTM outputs) (ablation) | TAAS · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Improved |
| add word-level attentive sub-network (vs pre-trained sent2vec sentence vectors) (ablation) | TAAS-KS · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Improved |
| initialize GRU with title representation instead of random (ablation) | TAAS-A · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Improved |
| add key-value memory network on top of GRU (ablation) | TAAS-G · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Improved |
| add text content (title+abstract) to interaction-only BPR | BPR · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Improved |
| use CNN item encoder (ConvMF) instead of auto-encoder (CDL) | content-based CF · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Improved |
| increase semantic weights alpha, beta beyond (0.4, 0.6) (ablation) | TAAS · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Hurt |
| model title-abstract semantic relationship (TAAS vs ConvMF, CDL, BPR) | paper recommendation · citeulike-a and PRSDataset | Pre@n, MRR, NDCG | Improved |
Closest papers by meaning
- TopicRNN: A Recurrent Neural Network with Long-Range Semantic Dependency (ICLR 2017 · similarity 0.85)
- Meta-Transfer Learning for Low-Resource Abstractive Summarization (AAAI 2021 · similarity 0.85)
- MeanSum: A Neural Model for Unsupervised Multi-Document Abstractive Summarization (ICML 2019 · similarity 0.85)
- SummaRuNNer: A Recurrent Neural Network Based Sequence Model for Extractive Summarization of Documents (AAAI 2017 · similarity 0.85)
- Exploring Explainable Selection to Control Abstractive Summarization (AAAI 2021 · similarity 0.84)
- Asymmetrical Hierarchical Networks with Attentive Interactions for Interpretable Review-Based Recommendation (AAAI 2020 · similarity 0.84)
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.
