Research map / Large language models / Retrieval and knowledge
Thinking Globally, Acting Locally: Distantly Supervised Global-to-Local Knowledge Selection for Background Based Conversation
Pengjie Ren, Zhumin Chen, Christof Monz, Jun Ma, Maarten de Rijke
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 global knowledge selection (GKS) to local knowledge selection (ablation) | GLKS · Holl-E mixed-short background (256-word), SR | ROUGE-1: 41.80 -> 44.52 | Improved |
| Add distant-supervision loss (ablation) | GLKS · Holl-E mixed-short background (256-word), SR | ROUGE-1: 41.27 -> 44.52 | Improved |
| Add MCE loss (ablation) | GLKS · Holl-E mixed-short background (256-word), SR | ROUGE-1: 43.69 -> 44.52 | Improved |
| Use GLKS rather than RefNet | Background-based conversation generation · Holl-E oracle background (256-word), SR | ROUGE-1: 42.87 -> 43.75 | Improved |
| Use GLKS rather than RefNet | Background-based conversation generation · Holl-E mixed-short background (256-word), SR | ROUGE-1: 41.33 -> 44.52 | Improved |
| Use GLKS rather than RefNet | Background-based conversation generation · Holl-E mixed-long background (1,200-word), SR | ROUGE-1: 35.30 -> 34.90 | Hurt |
| Increase the unfolding window size from m ≤ 2 to m ≥ 3 (ablation) | GLKS · Holl-E | ROUGE | Improved |
| Vary the unfolding window size m (ablation) | GLKS · Holl-E | ROUGE: around 2% | Mixed |
Closest papers by meaning
- Sequential Latent Knowledge Selection for Knowledge-Grounded Dialogue (ICLR 2020 · similarity 0.88)
- A Knowledge-Grounded Neural Conversation Model (AAAI 2018 · similarity 0.87)
- PRAGWORLD: A Benchmark Evaluating LLMs’ Local World Model Under Minimal Linguistic Alterations and Conversational Dynamics (AAAI 2026 · similarity 0.87)
- Zero-Resource Knowledge-Grounded Dialogue Generation (NeurIPS 2020 · similarity 0.87)
- RefNet: A Reference-Aware Network for Background Based Conversation (AAAI 2020 · similarity 0.87)
- Improving Open-Domain Dialogue Response Generation with Multi-Source Multilingual Commonsense Knowledge (AAAI 2024 · 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.
