Research map / Large language models / Retrieval and knowledge
Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation
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 goal-related reward factors (coherence, user interest, diversity, sustainability, goal completion) (ablation) | KnowHRL · DuConv with user simulator | inter-topic coherence: 0.41 -> 1.39 | Helped |
| add goal-related reward factors (ablation) | KnowHRL · DuConv with user simulator | user-interest consistency: 0.07 -> 0.89 | Helped |
| add goal-related reward factors (ablation) | KnowHRL · DuConv with user simulator | intra-topic coherence: 1.13 -> 1.42 | Helped |
| add goal-related reward factors (ablation) | KnowHRL · DuConv with user simulator | knowledge accuracy: 0.81 -> 0.88 | Helped |
| add goal-related reward factors (ablation) | KnowHRL · DuConv with user simulator | Dist-2: 0.35 -> 0.39 | Helped |
| add goal-related reward factors (ablation) | KnowHRL · DuConv with human | inter-topic coherence: 0.46 -> 1.45 | Helped |
| hierarchical goal planning KnowHRL instead of CCM+LaRL | open-domain dialog model · DuConv with user simulator | inter-topic coherence: 0.44 -> 1.39 | Helped |
| KnowHRL instead of CCM+LaRL | open-domain dialog model · DuConv with user simulator | knowledge accuracy: 0.17 -> 0.88 | Helped |
| KnowHRL instead of CCM+LaRL | open-domain dialog model · DuConv turn-level, user simulator | appropriateness: 0.77 -> 0.87 | Helped |
| add RL (LaRL) on top of CCM | CCM · DuConv with user simulator | intra-topic coherence: 0.67 -> 0.94 | Helped |
Closest papers by meaning
- HiTKG: Towards Goal-Oriented Conversations via Multi-Hierarchy Learning (AAAI 2022 · similarity 0.92)
- Wizard of Wikipedia: Knowledge-Powered Conversational Agents (ICLR 2019 · similarity 0.90)
- Hierarchical Reinforcement Learning for Open-Domain Dialog (AAAI 2020 · similarity 0.89)
- Hierarchical Text Generation and Planning for Strategic Dialogue (ICML 2018 · similarity 0.89)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.87)
- Simulation-Free Hierarchical Latent Policy Planning for Proactive Dialogues (AAAI 2025 · similarity 0.87)
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.