Research map / Vision-language and multimodal / Human and embodied modeling
Knowledge Enhanced Hybrid Neural Network for Text Matching
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 prior knowledge and combine word, sequential-structure, and knowledge-enhanced matching channels | text matching · SemEval 2015 answer selection test | ACC: 0.736 -> 0.748 | Helped |
| add prior knowledge and combine word, sequential-structure, and knowledge-enhanced matching channels | text matching · Ubuntu response selection test | R2@1: 0.767 -> 0.786 | Helped |
| add prior knowledge and combine word, sequential-structure, and knowledge-enhanced matching channels | text matching · Ubuntu response selection test | R10@1: 0.420 -> 0.460 | Helped |
| add prior knowledge and combine word, sequential-structure, and knowledge-enhanced matching channels | text matching · Ubuntu response selection test | R10@2: 0.565 -> 0.591 | Helped |
| add prior knowledge and combine word, sequential-structure, and knowledge-enhanced matching channels | text matching · Ubuntu response selection test | R10@5: 0.801 -> 0.819 | Helped |
| add sequential-structure and knowledge-enhanced channels to word matching (ablation) | KEHNN · SemEval 2015 answer selection test | ACC: 0.717 -> 0.748 | Helped |
| add word and knowledge-enhanced channels to sequential-structure matching (ablation) | KEHNN · SemEval 2015 answer selection test | ACC: 0.734 -> 0.748 | Helped |
| add word and sequential-structure channels to knowledge-enhanced matching (ablation) | KEHNN · SemEval 2015 answer selection test | ACC: 0.738 -> 0.748 | Helped |
| add sequential-structure and knowledge-enhanced channels to word matching (ablation) | KEHNN · Ubuntu response selection | R2@1: 0.743 -> 0.786 | Helped |
| add word and knowledge-enhanced channels to sequential-structure matching (ablation) | KEHNN · Ubuntu response selection | R2@1: 0.779 -> 0.786 | Helped |
| add word and sequential-structure channels to knowledge-enhanced matching (ablation) | KEHNN · Ubuntu response selection | R2@1: 0.750 -> 0.786 | Helped |
| use the full model rather than LSTM | KEHNN · QA text pairs length [60,90) | accuracy: 0.728 -> 0.765 | Helped |
Closest papers by meaning
- Deconvolutional Latent-Variable Model for Text Sequence Matching (AAAI 2018 · similarity 0.86)
- Improved Text Matching by Enhancing Mutual Information (AAAI 2018 · similarity 0.86)
- Knowledge and Cross-Pair Pattern Guided Semantic Matching for Question Answering (AAAI 2020 · similarity 0.86)
- LET: Linguistic Knowledge Enhanced Graph Transformer for Chinese Short Text Matching (AAAI 2021 · similarity 0.85)
- Identifying Query-Relevant Neurons in Large Language Models for Long-Form Texts (AAAI 2025 · similarity 0.83)
- Improved Graph Contrastive Learning for Short Text Classification (AAAI 2024 · 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.