ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation

• 类型：arxiv
• 标识：2609.09076
• 链接：https://arxiv.org/abs/2609.09076
• 主分类：multimodal
• 形态：method
• TLDR：As LLMs are increasingly used for pre-submission self-review, there is growing demand for feedback that not only identifies weaknesses but also guides authors toward concrete revisions. We study this as Actionable Peer-review Generation and decompose it into two subtasks: diagnostic claim generation and revision suggestion generation. We introduce ActReview, a rebuttal-guided post-training framework that connects paper-specific diagnoses to concrete, grounded revision plans. Our central insight is that author rebuttals reveal plausible actions for addressing reviewer concerns and can therefore
• 副分类：engineering
• 待LLM分类：否
• 标题中文：ActReview：基于反驳引导训练数据与评分量规奖励的可操作同行评审生成
• TLDR中文：随着 LLM 越来越多地用于投稿前自我审稿，对于不仅能识别弱点还能指导作者进行具体修改的反馈需求日益增长。我们将此问题研究为可操作同行评审生成（Actionable Peer-review Generation），并将其分解为两个子任务：诊断性主张生成与修改建议生成。我们提出 ActReview，一个基于反驳引导的后训练框架，将论文级诊断与具体、可落地的修改方案相连接。我们的核心洞察是：作者的反驳揭示了应对审稿人关切的可行操作，因此
• 来源文件：
• /inbox/tom/_candidates/2026-09-12-agent-rag-longcontext-candidates.json