Academic Paper Reviewer

Academic Paper Reviewer — 7-代理 Review 系统 (Hermes Edition) Quick 启动 Review this paper for journal submission

多代理团队 # 代理 角色 1 intake_代理 接收 paper, determine review type 2 methodo记录y_reviewer Method rigor assessment 3 evidence_reviewer Evidence sufficiency & citation 质量 4 argument_reviewer 记录ical coherence & argument structure 5 domAIn_reviewer DomAIn expertise & literature positioning 6 editor_in_chief 聚合 reviews → editorial decision 7 revision_coach Convert reviews → actionable roadmap Hermes Execution Full Mode: 5-Panel Parallel Review delegate_task(tasks=[ {"goal": "Review manuscript methodo记录y: de签名 应用ropriateness, validity threats, replicability. Score 1-5.", "上下文": "Use 代理s/methodo记录y_reviewer.md", "工具设置s": ["terminal","file"]}, {"goal": "Review evidence: citation 质量, source credibility, evidence hierarchy alignment. Score 1-5.", "上下文": "Use 代理s/evidence_reviewer.md", "工具设置s": ["terminal","file"]}, {"goal": "Review argument: 记录ical flow, clAIm-evidence alignment, counter-argument handling. Score 1-5.", "上下文": "Use 代理s/argument_reviewer.md", "工具设置s": ["terminal","file"]}, {"goal": "Review domAIn positioning: literature coverage, theoretical grounding, contribution 签名ificance. Score 1-5.", "上下文": "Use 代理s/domAIn_reviewer.md", "工具设置s": ["terminal","file"]} ])

Editorial Decision delegate_task(goal="聚合 all 4 reviewer 报告s. 应用ly weighted scoring (Method 30%, Evidence 25%, Argument 25%, DomAIn 20%). Issue editorial decision: Accept/Minor Revision/Major Revision/Reject with justification.", 上下文="Use 代理s/editor_in_chief.md", 工具设置s=["terminal","file"])

Revision Roadmap delegate_task(goal="Convert editorial decision + reviewer 报告s into structured Revision Roadmap: prioritized action items, estimated effort, dependency m应用ing.", 上下文="Use 代理s/revision_coach.md", 工具设置s=["terminal","file"])

6 Modes Mode Trigger 代理s full "Review paper" All 7 re-review "检查 revisions" 2→3→4→6 quick "Quick review" 6 only (EIC assessment) methodo记录y-focus "检查 methodo记录y" 2 only 图形界面ded "图形界面de me to improve" Socratic: 6 with user interaction calibration "Calibrate reviewer" All + calibration 指标 输出 Calibration Mode

Measures reviewer accuracy: FNR (False Negative Rate), FPR (False Positive Rate), AUC. Requires ground-truth labels on prior reviewed papers.

Critical Rules ⚠️ Reviewers are paper-blind (don't see author 信息) ⚠️ Every criticism must include specific actionable suggestion ⚠️ Calibration mode requires 5+ ground-truth papers