Research map / Large language models / Retrieval and knowledge
Automatic Paper Reviewing with Heterogeneous Graph Reasoning over LLM-Simulated Reviewer-Author Debates
Shuaimin Li, Liyang Fan, Yufang Lin, Zeyang Li, Xian Wei, Shiwen Ni, Hamid Alinejad-Rokny, Min Yang
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 title node (ablation) | ReViewGraph · ICLR 2025 | accuracy: 66.49 -> 71.65 | Improved |
| add title node (ablation) | ReViewGraph · ICLR 2023 | accuracy: 67.03 -> 70.29 | Improved |
| add title node (ablation) | ReViewGraph · ICLR 2024 | accuracy: 64.73 -> 66.10 | Improved |
| add evaluation dimension nodes (ablation) | ReViewGraph · ICLR 2025 | accuracy: 68.30 -> 71.65 | Improved |
| add reviewer-author interaction edges (ablation) | ReViewGraph · ICLR 2025 | accuracy: 70.36 -> 71.65 | Improved |
| add reviewer-author interaction edges (ablation) | ReViewGraph · ICLR 2023 | accuracy: 69.57 -> 70.29 | Improved |
| add inter-reviewer relation edges (ablation) | ReViewGraph · ICLR 2025 | accuracy: 69.59 -> 71.65 | Improved |
| use heterogeneous graph instead of homogeneous (ablation) | ReViewGraph · ICLR 2025 | accuracy: 70.62 -> 71.65 | Improved |
| use heterogeneous graph instead of homogeneous (ablation) | ReViewGraph · ICLR 2024 | accuracy: 65.75 -> 66.10 | No clear effect |
| ReViewGraph vs CycleReviewer-70B (fine-tuned) baseline | ReViewGraph · ICLR 2025 macro F1 | accuracy: 59.26 -> 71.65 | Improved |
| ReViewGraph vs ICL prompting baseline | ReViewGraph · ICLR 2023 | accuracy: 58.33 -> 70.29 | Improved |
| debate-graph model vs DeepReview-14B-Std | ReViewGraph · ICLR 2024 | accuracy: 59.93 -> 66.10 | Improved |
Closest papers by meaning
- When Debate Helps: Proposal Supply and Verification-Aware Readout in Multi-Agent Reasoning (NeurIPS 2026 · similarity 0.86)
- Debate on Graph: A Flexible and Reliable Reasoning Framework for Large Language Models (AAAI 2025 · similarity 0.86)
- Mixture of Knowledge Minigraph Agents for Literature Review Generation (AAAI 2025 · similarity 0.86)
- Precise but Uncoupled: Reviewer Precision Does Not Guarantee Critique Uptake in Multi-Agent Math Reasoning (NeurIPS 2026 · similarity 0.86)
- Multi-Agent Debate for LLM Judges with Adaptive Stability Detection (NeurIPS 2025 · similarity 0.85)
- Agent Reviewers: Domain-specific Multimodal Agents with Shared Memory for Paper Review (ICML 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.
