Research map / Label-efficient and robust learning / Domain adaptation and generalization
Dynamic Knowledge Routing Network for Target-Guided Open-Domain Conversation
Jinghui Qin, Zheng Ye, Jianheng Tang, Xiaodan Liang
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 |
|---|---|---|---|
| Use DKRN with keyword relation routing and dual target-guided selection instead of Kernel | Target-guided dialogue on TGPC · TGPC self-play | Succ. (%): 62.56 -> 89.0 | Improved |
| Use DKRN with keyword relation routing and dual target-guided selection instead of Kernel | Target-guided dialogue on CWC · CWC self-play | Succ. (%): 53.2 -> 84.4 | Improved |
| Use DKRN instead of Kernel | Dialogue agent · CWC human rating | Smoothness: 2.48 -> 3.22 | Improved |
| Use DKRN instead of Kernel | Dialogue agent · CWC human rating | Succ. (%): 58.0 -> 88.0 | Improved |
| Use DKRN instead of Neural | Keyword prediction · TGPC | Rw@1: 0.0708 -> 0.0909 | Improved |
| Use DKRN instead of Neural | Keyword prediction · CWC | Rw@1: 0.0654 -> 0.0680 | Improved |
| Use DKRN instead of Kernel | Response retrieval · TGPC | MRR: 0.6877 -> 0.7110 | Improved |
| Use DKRN instead of Neural | Response retrieval · CWC | MRR: 0.7326 -> 0.7533 | Improved |
| Use DKRN instead of Kernel | Target-guided dialogue · TGPC self-play | #Turns: 4.65 -> 5.02 | Hurt |
Closest papers by meaning
- Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation (AAAI 2020 · similarity 0.88)
- HiTKG: Towards Goal-Oriented Conversations via Multi-Hierarchy Learning (AAAI 2022 · similarity 0.88)
- Keyword-Guided Neural Conversational Model (AAAI 2021 · similarity 0.88)
- CORECODE: A Common Sense Annotated Dialogue Dataset with Benchmark Tasks for Chinese Large Language Models (AAAI 2024 · similarity 0.85)
- Improving Open-Domain Dialogue Response Generation with Multi-Source Multilingual Commonsense Knowledge (AAAI 2024 · similarity 0.84)
- Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training (ICLR 2025 · 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.
