Machine Learning
Multi-Round Academic Paper Peer Review
Multi-round academic paper peer review simulation using multi-model diversity for improving paper quality before submission. Assigns different LLM models (Claude, GPT/Codex, Gemini) to 4 specialized reviewer roles (security domain expert, ML/GNN methods expert, experimental methodology expert, writing/presentation expert) plus a meta-reviewer and devil's advocate. Covers technical soundness, novelty, ML-for-security pitfalls, experimental rigor, and presentation quality. Supports multi-round cross-model review-rebuttal-revision cycles. Use when the user wants paper review, peer review simulation, technical feedback, paper improvement, strength/weakness analysis, or mentions reviewing, referee, rebuttal, or revision.