Research map / Large language models / Language, safety and interpretability
WikiREVIEW: A Multi-Perspective Review Framework for Automatic Wiki-Style Article Generation
Guo-Biao Zhang, Zhijing Wu, Tian Lan, Ding-Yuan Liu, Yu-Shi Zhu, Xian-Ling Mao
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 multi-perspective review after initial generation (WikiREVIEW vs OmniThink) | wiki article generation · FreshWiki (GPT-4o) | Relevance (1-5): 4.40 -> 4.65 | Improved |
| add multi-perspective review (WikiREVIEW vs OmniThink) | wiki article generation · ChineseWiki (DeepSeek-R1) | Relevance (1-5): 4.47 -> 4.63 | Improved |
| add multi-perspective review (WikiREVIEW vs OmniThink) | wiki article generation · FreshWiki (GPT-4o) | ROUGE-1: 54.51 -> 54.52 | No clear effect |
| add multi-perspective review (WikiREVIEW vs OmniThink) | wiki article generation · FreshWiki (Qwen-2.5) | ROUGE-1: 51.45 -> 54.12 | Improved |
| remove multi-perspective review mechanism (ablation) | WikiREVIEW (GPT-4o) · FreshWiki | Rubric grading (1-5) | Improved |
| add review mechanism (outline quality) | wiki article generation · FreshWiki outline quality | Logical Coherence (1-5): 3.97 -> 4.01 | Improved |
| add multi-perspective review (human eval) | wiki article generation · FreshWiki human eval (20 topics) | Relevance (7-point): 5.1 -> 5.8 | Improved |
| add multi-perspective review (human eval) | wiki article generation · FreshWiki human eval (20 topics) | Interest Level (7-point): 4.9 -> 5.1 | Mixed |
| add credibility-perspective review | wiki article generation · FreshWiki RAGAs | Faithfulness: 88.52 -> 91.71 | Improved |
| add review mechanism | wiki article generation · FreshWiki RAGAs | Context Relevance: 78.28 -> 78.96 | No clear effect |
Closest papers by meaning
- WikiSQE: A Large-Scale Dataset for Sentence Quality Estimation in Wikipedia (AAAI 2024 · similarity 0.85)
- EMERGE: A Benchmark for Updating Knowledge Graphs with Emerging Textual Knowledge (NeurIPS 2026 · similarity 0.82)
- WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia (NeurIPS 2024 · similarity 0.82)
- Chain-of-Thought Improves Text Generation with Citations in Large Language Models (AAAI 2024 · similarity 0.82)
- WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs (ICML 2025 · similarity 0.82)
- Retrieval is Accurate Generation (ICLR 2024 · 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.
