Research map / Label-efficient and robust learning / Recognition and long-tail
Semantic Parsing in Task-Oriented Dialog with Recursive Insertion-Based Encoder
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 |
|---|---|---|---|
| replace the decoder-based Seq2seq-Ptr parser with recursive insertion-based RINE | TOP test, RoBERTa-large · TOP | exact match: 87.10 -> 87.57 | Helped |
| use RINE rather than the strongest reported Seq2seq-Ptr baseline | TOPv2 Reminder 25 SPIS · TOPv2 Reminder 25 SPIS | exact match: 55.7 -> 71.10$\pm$0.63 | Helped |
| use RINE rather than the strongest reported Seq2seq-Ptr baseline | TOPv2 Reminder 500 SPIS · TOPv2 Reminder 500 SPIS | exact match: 71.9 -> 81.31$\pm$0.22 | Helped |
| use RINE rather than the strongest reported Seq2seq-Ptr baseline | TOPv2 Weather 25 SPIS · TOPv2 Weather 25 SPIS | exact match: 71.6 -> 77.03$\pm$0.16 | Helped |
| use RINE rather than the strongest reported Seq2seq-Ptr baseline | TOPv2 Weather 500 SPIS · TOPv2 Weather 500 SPIS | exact match: 84.9 -> 87.80$\pm$0.04 | Helped |
| replace BERT-MRC with RINE | ACE2005 · ACE2005 | F1: 86.88 -> 86.44$\pm$0.04 | Hurt |
| use top-down instead of bottom-up generation order (ablation) | RINE with RoBERTa-base · TOP validation | exact match: 87.01$\pm$0.04 -> 87.07$\pm$0.09 | No clear effect |
| use top-down instead of bottom-up generation order (ablation) | RINE with RoBERTa-large · TOP validation | exact match: 87.70$\pm$0.04 -> 87.72$\pm$0.07 | No clear effect |
| use RINE rather than Seq2seq-Ptr for inference | TOP validation with RoBERTa-base · TOP validation | sentences per second: 3.70 -> 12.95 | Helped |
Closest papers by meaning
- Dialog-to-Action: Conversational Question Answering Over a Large-Scale Knowledge Base (NeurIPS 2018 · similarity 0.86)
- Guiding Attention in Sequence-to-Sequence Models for Dialogue Act Prediction (AAAI 2020 · similarity 0.86)
- GraphMemDialog: Optimizing End-to-End Task-Oriented Dialog Systems Using Graph Memory Networks (AAAI 2022 · similarity 0.86)
- Dynamic Hybrid Relation Exploration Network for Cross-Domain Context-Dependent Semantic Parsing (AAAI 2021 · similarity 0.85)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.85)
- From Retrieval to Generation: A Simple and Unified Generative Model for End-to-End Task-Oriented Dialogue (AAAI 2024 · 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.