Research map / Large language models / Agents, code and math
MALA: Cross-Domain Dialogue Generation with Action Learning
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 semantic latent actions learned with state transitions instead of lexical-similarity latent actions | MALA-S3 vs LaRL · MultiWOZ joint training | Entity-F1: 71.3 -> 76.8 | Helped |
| Add action alignment across domains (Stage-II) after Stage-I (ablation) | MALA · MultiWOZ Attraction cross-domain generation | Entity-F1: 31.5 -> 36.5 | Helped |
| Add domain-specific action learning with SPN (Stage-III) after Stage-II (ablation) | MALA · MultiWOZ Attraction cross-domain generation | Entity-F1: 36.5 -> 41.9 | Helped |
| Add pointwise state-transition loss (ablation) | MALA semantic action learning · SMD joint training, 55% state annotations | Entity-F1: 5.9% | Helped |
| Add pairwise state-transition loss (ablation) | MALA semantic action learning · SMD joint training, 55% state annotations | Entity-F1: 8.0% | Helped |
| Use lexical-similarity latent actions with fine-tuning rather than MALA's cross-domain semantic action learning | LaRL · MultiWOZ Attraction cross-domain generation | Entity-F1: less than 3% improvement | No clear effect |
Closest papers by meaning
- Modelling Hierarchical Structure between Dialogue Policy and Natural Language Generator with Option Framework for Task-oriented Dialogue System (ICLR 2021 · similarity 0.88)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.86)
- From Retrieval to Generation: A Simple and Unified Generative Model for End-to-End Task-Oriented Dialogue (AAAI 2024 · similarity 0.86)
- Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training (ICLR 2025 · similarity 0.85)
- On Discriminative vs. Generative classifiers: Rethinking MLLMs for Action Understanding (ICLR 2026 · similarity 0.85)
- Simulation-Free Hierarchical Latent Policy Planning for Proactive Dialogues (AAAI 2025 · 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.