Research map / Reinforcement learning / Visual and representation RL
Dialog State Tracking with Reinforced Data Augmentation
Yichun Yin, Lifeng Shang, Xin Jiang, Xiao Chen, Qun Liu
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 |
|---|---|---|---|
| Add reinforced data augmentation | GLAD* · WoZ | joint goal accuracy: 88.3 -> 90.7 | Improved |
| Add reinforced data augmentation | GLAD* · MultiWoZ (restaurant) | joint goal accuracy: 83.6 -> 86.7 | Improved |
| Add reinforced data augmentation | NBT-CNN · WoZ | joint goal accuracy: 84.0 -> 87.9 | Improved |
| Add reinforced data augmentation | NBT-CNN · MultiWoZ (restaurant) | joint goal accuracy: 79.8 -> 83.4 | Improved |
| Apply coarse-grained random data augmentation | GLAD* · WoZ | joint goal accuracy: 88.3 -> 88.0 | Hurt |
| Apply coarse-grained random data augmentation | GLAD* · MultiWoZ (restaurant) | joint goal accuracy: 83.6 -> 82.7 | Hurt |
| Add RDA using 10% of training data | GLAD* · WoZ, 10% training data | joint goal accuracy: 50.1 -> 66.8 | Improved |
| Add RDA using 10% of training data | GLAD* · MultiWoZ (restaurant), 10% training data | joint goal accuracy: 60.0 -> 71.5 | Improved |
| Remove bag-level reward; equivalently, add bag-level reward (ablation) | GLAD* + RDA · WoZ | joint goal accuracy: 89.1 -> 90.7 | Improved |
| Remove bag-level reward; equivalently, add bag-level reward (ablation) | GLAD* + RDA · MultiWoZ (restaurant) | joint goal accuracy: 84.3 -> 86.7 | Improved |
| Remove instance-level reward; equivalently, add instance-level reward (ablation) | GLAD* + RDA · WoZ | joint goal accuracy: 89.8 -> 90.7 | Improved |
| Remove instance-level reward; equivalently, add instance-level reward (ablation) | GLAD* + RDA · MultiWoZ (restaurant) | joint goal accuracy: 85.4 -> 86.7 | Improved |
Closest papers by meaning
- Efficient Dialog Policy Learning by Reasoning with Contextual Knowledge (AAAI 2022 · similarity 0.87)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.84)
- Memory-T1: Reinforcement Learning for Temporal Reasoning in Multi-session Agents (ICLR 2026 · similarity 0.82)
- An Efficient Self-Learning Framework For Interactive Spoken Dialog Systems (ICML 2024 · similarity 0.82)
- Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents (ICLR 2024 · similarity 0.82)
- TTRL: Test-Time Reinforcement Learning (NeurIPS 2025 · similarity 0.82)
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.
