Research map / Label-efficient and robust learning / Recognition and long-tail
Duplicate Question Identification by Integrating FrameNet With Neural Networks
Xiaodong Zhang, Xu Sun, Houfeng Wang
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 |
|---|---|---|---|
| Ensemble Frame-GBDT with AI-BLSTM | AI-BLSTM · Quora Question Pairs | accuracy: 87.43 -> 88.53 | Improved |
| Ensemble Frame-GBDT with BLSTM | BLSTM · Quora Question Pairs | accuracy: 86.17 -> 87.92 | Improved |
| Ensemble Frame-GBDT with BM25 | BM25 · Quora Question Pairs | accuracy: 65.23 -> 86.12 | Improved |
| Add frame embeddings to BLSTM input (embedding approach) | BLSTM · Quora Question Pairs | accuracy: 86.17 -> 86.42 | Improved |
| Add frame embeddings to AI-BLSTM input | AI-BLSTM · Quora Question Pairs | accuracy: 87.43 -> 87.62 | Improved |
| Ensemble BLSTM with Frame-GBDT instead of with CNN (ablation) | BLSTM ensemble · Quora Question Pairs | accuracy: 87.14 -> 87.92 | Improved |
| Ensemble BLSTM with Frame-GBDT instead of with AP-BLSTM (ablation) | BLSTM ensemble · Quora Question Pairs | accuracy: 87.01 -> 87.92 | Improved |
| Ensemble weight alpha near 0.5 versus extremes (ablation) | BLSTM + Frame-GBDT · Quora Question Pairs | accuracy | Improved |
Closest papers by meaning
- An Unsupervised Model With Attention Autoencoders for Question Retrieval (AAAI 2018 · similarity 0.84)
- Knowledge and Cross-Pair Pattern Guided Semantic Matching for Question Answering (AAAI 2020 · similarity 0.84)
- Question-Answering with Grammatically-Interpretable Representations (AAAI 2018 · similarity 0.84)
- A Compare-Aggregate Model for Matching Text Sequences (ICLR 2017 · similarity 0.84)
- Improved Text Matching by Enhancing Mutual Information (AAAI 2018 · similarity 0.83)
- Neural Semantic Parsing in Low-Resource Settings with Back-Translation and Meta-Learning (AAAI 2020 · similarity 0.83)
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.
