Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
MA-DST: Multi-Attention-Based Scalable Dialog State Tracking
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 |
|---|---|---|---|
| Replace TRADE with MA-DST | Multi-domain DST · MultiWOZ 2.1 multi-domain | Joint goal accuracy: 45.6 -> 51.04 | Helped |
| Add slot gate and word-level cross-attention to the base model (ablation) | MA-DST · MultiWOZ 2.1 multi-domain | Joint goal accuracy: 44.0 -> 47.60 | Helped |
| Add higher-level cross-attention (ablation) | MA-DST with slot gate and word-level cross-attention · MultiWOZ 2.1 multi-domain | Joint goal accuracy: 47.60 -> 49.56 | Helped |
| Add self-attention and slot summarizer (ablation) | MA-DST with higher-level cross-attention · MultiWOZ 2.1 multi-domain | Joint goal accuracy: 49.56 -> 50.55 | Helped |
| Add ELMo embeddings (ablation) | MA-DST with self-attention and slot summarizer · MultiWOZ 2.1 multi-domain | Joint goal accuracy: 50.55 -> 51.04 | Helped |
| Ensemble MA-DST models | MA-DST · MultiWOZ 2.1 multi-domain | Joint goal accuracy: 51.04 -> 51.88 | Helped |
| Replace TRADE with MA-DST | Single-domain DST · MultiWOZ 2.1 single-domain Hotel | Joint goal accuracy: 50.25 -> 57.70 | Helped |
| Replace TRADE with MA-DST | Single-domain DST · MultiWOZ 2.1 single-domain Train | Joint goal accuracy: 74.47 -> 76.47 | Helped |
| Replace TRADE with MA-DST | Single-domain DST · MultiWOZ 2.1 single-domain Taxi | Joint goal accuracy: 70.18 -> 76.55 | Helped |
| Replace TRADE with MA-DST | Single-domain DST · MultiWOZ 2.1 single-domain Restaurant | Joint goal accuracy: 66.02 -> 66.33 | Helped |
| Replace TRADE with MA-DST | Single-domain DST · MultiWOZ 2.1 single-domain Attraction | Joint goal accuracy: 68.48 -> 72.49 | Helped |
| Replace TRADE with MA-DST | Zero-shot DST · MultiWOZ 2.1 zero-shot Taxi | Joint goal accuracy: 59.21 -> 59.27 | No clear effect |
Closest papers by meaning
- Non-Autoregressive Dialog State Tracking (ICLR 2020 · similarity 0.92)
- Learning to Imagine: Distillation-Based Interactive Context Exploitation for Dialogue State Tracking (AAAI 2023 · similarity 0.90)
- Schema-Guided Multi-Domain Dialogue State Tracking with Graph Attention Neural Networks (AAAI 2020 · similarity 0.90)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.87)
- Structure-Aware Multimodal Sequential Learning for Visual Dialog (AAAI 2024 · similarity 0.86)
- StateTree: Enhancing Long-term Dialogue Reasoning via Reinforcement Learning (NeurIPS 2026 · similarity 0.85)
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.