Research map / Reinforcement learning / Visual and representation RL
Modelling Hierarchical Structure between Dialogue Policy and Natural Language Generator with Option Framework for Task-oriented Dialogue System
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 asynchronous rather than synchronous updates for dialogue policy and NLG (ablation) | HDNO · MultiWoz 2.0 | Total: 98.17 -> 109.40 | Helped |
| use asynchronous rather than synchronous updates for dialogue policy and NLG (ablation) | HDNO · MultiWoz 2.1 | Total: 95.76 -> 106.77 | Helped |
| add hierarchical reinforcement learning after Bayesian pretraining (ablation) | HDNO · MultiWoz 2.0 | Total: 84.85 -> 109.40 | Helped |
| add hierarchical reinforcement learning after Bayesian pretraining (ablation) | HDNO · MultiWoz 2.1 | Total: 86.22 -> 106.77 | Helped |
| add a language-model discriminator reward alongside success reward (ablation) | HDNO · MultiWoz 2.0 | Total: 108.66 -> 109.40 | Helped |
| replace discriminator reward with BLEU reward alongside success reward (ablation) | HDNO · MultiWoz 2.0 | Total: 109.40 -> 108.44 | Hurt |
| vary discriminator reward weight within the tested range (ablation) | HDNO · MultiWoz 2.0 | Total: 109.40; 109.10; 109.24; 109.27 | No clear effect |
| model dialogue acts as options and words as primitive actions | HDNO · MultiWoz 2.0 | Total: 101.25 -> 109.40 | Helped |
| use asynchronous hierarchical policy optimization rather than word-level E2E RL | HDNO · MultiWoz 2.0 | Total: 102.52 -> 109.40 | Helped |
| use asynchronous hierarchical policy optimization rather than supervised HDSA | HDNO · MultiWoz 2.0 | Total: 104.30 -> 109.40 | Helped |
Closest papers by meaning
- Simulation-Free Hierarchical Latent Policy Planning for Proactive Dialogues (AAAI 2025 · similarity 0.90)
- Dialogue for Prompting: A Policy-Gradient-Based Discrete Prompt Generation for Few-Shot Learning (AAAI 2024 · similarity 0.87)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.87)
- Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training (ICLR 2025 · similarity 0.87)
- Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents (ICLR 2024 · similarity 0.87)
- Natural Language Actor-Critic: Scalable Off-Policy Learning in Language Space (NeurIPS 2026 · 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.