Research map / Architectures and efficiency / RNNs and normalization
Global-to-local Memory Pointer Networks for Task-Oriented Dialogue
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 GLMP K1 instead of Mem2Seq | bAbI T5 OOV · bAbI T5 OOV | per-response accuracy: 84.5 -> 92.0 | Helped |
| use GLMP K1 instead of Mem2Seq | bAbI T5 OOV · bAbI T5 OOV | task-completion rate: 2.3 -> 21.7 | Helped |
| use GLMP K3 instead of KVR* | SMD · SMD | BLEU: 13.2 -> 14.79 | Helped |
| use GLMP K3 instead of KVR* | SMD · SMD | Entity F1: 48.0 -> 59.97 | Helped |
| use GLMP K1 rather than Mem2Seq | SMD human evaluation · SMD | Appropriate: 3.89 -> 4.15 | Helped |
| use GLMP K1 rather than Mem2Seq | SMD human evaluation · SMD | Humanlike: 3.80 -> 4.02 | Helped |
| add dialogue-history hidden-state writing to external memory (ablation) | GLMP K1 · bAbI T5 OOV | per-response accuracy: 86.2 -> 92.0 | Helped |
| add dialogue-history hidden-state writing to external memory (ablation) | GLMP K1 · SMD | Entity F1: 47.96 -> 57.25 | Helped |
| add dialogue-history hidden-state writing to external memory (ablation) | GLMP K1 · bAbI T1 OOV | per-response accuracy: 90.4 -> 100 | Helped |
| add dialogue-history hidden-state writing to external memory (ablation) | GLMP K1 · bAbI T2 OOV | per-response accuracy: 85.6 -> 100 | Helped |
| add global memory pointer weighting of external memory (ablation) | GLMP K1 · SMD | Entity F1: 45.78 -> 57.25 | Helped |
| add global memory pointer weighting of external memory (ablation) | GLMP K1 · bAbI T2 OOV | per-response accuracy: 91.7 -> 100 | Helped |
Closest papers by meaning
- From Retrieval to Generation: A Simple and Unified Generative Model for End-to-End Task-Oriented Dialogue (AAAI 2024 · similarity 0.87)
- StreamingDialogue: Prolonged Dialogue Learning via Long Context Compression with Minimal Losses (NeurIPS 2024 · similarity 0.87)
- From Isolated Conversations to Hierarchical Schemas: Dynamic Tree Memory Representation for LLMs (ICLR 2025 · similarity 0.86)
- Compressed Context Memory for Online Language Model Interaction (ICLR 2024 · similarity 0.86)
- StateTree: Enhancing Long-term Dialogue Reasoning via Reinforcement Learning (NeurIPS 2026 · similarity 0.85)
- Towards General Continuous Memory for Vision-Language Models (NeurIPS 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.