Research map / Label-efficient and robust learning / Recognition and long-tail
Dialogue Act Sequence Labeling Using Hierarchical Encoder With CRF
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 |
|---|---|---|---|
| Use hierarchical Bi-LSTM encoding and CRF label decoding | Dialogue act sequence labeling · SwDA | accuracy: 77.0 -> 79.2; 2.2% absolute points | Helped |
| Use hierarchical Bi-LSTM encoding and CRF label decoding | Dialogue act sequence labeling · MRDA | accuracy: 86.8 -> 90.9; 4.1% | Helped |
| Add an utterance encoder (ablation) | WE with LR · SwDA | accuracy: 71.4 -> 72.2 | Helped |
| Add an utterance encoder (ablation) | WE with CRF · SwDA | accuracy: 72.2 -> 72.7 | Helped |
| Add the conversation encoder (ablation) | WE+UL with LR · SwDA | accuracy: 72.2 -> 74.1 | Helped |
| Add the conversation encoder (ablation) | WE+UL with CRF · SwDA | accuracy: 72.7 -> 79.2 | Helped |
| Replace LR with CRF (ablation) | WE · SwDA | accuracy: 71.4 -> 72.2 | Helped |
| Replace LR with CRF (ablation) | WE+UL · SwDA | accuracy: 72.2 -> 72.7 | Helped |
| Replace LR with CRF (ablation) | WE+UL+CL · SwDA | accuracy: 74.1 -> 79.2 | Helped |
| Add POS tag embeddings and a parallel encoder (ablation) | Hierarchical Bi-LSTM-CRF · SwDA | accuracy: approximately 1%; 79.2 -> 77.9 | Hurt |
| Add intra-attention context over 10 previous utterances (ablation) | Hierarchical Bi-LSTM-CRF · SwDA | accuracy: 79.2 -> 77.4 | Hurt |
| Add intra-attention context over 5 previous utterances (ablation) | Hierarchical Bi-LSTM-CRF · SwDA | accuracy: 79.2 -> 78.3 | Hurt |
Closest papers by meaning
- Guiding Attention in Sequence-to-Sequence Models for Dialogue Act Prediction (AAAI 2020 · similarity 0.89)
- Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models (AAAI 2016 · similarity 0.88)
- DCR-Net: A Deep Co-Interactive Relation Network for Joint Dialog Act Recognition and Sentiment Classification (AAAI 2020 · similarity 0.87)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.86)
- ACT as Human: Multimodal Large Language Model Data Annotation with Critical Thinking (NeurIPS 2025 · similarity 0.84)
- Structure-Aware Multimodal Sequential Learning for Visual Dialog (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.