titleActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation
authorsYiling Ma, Yilun Zhao, Sihong Wu, Ziyu Chen, Manasi Patwardhan, Arman Cohan
date2026-09-08
venuearXiv (preprint, not peer reviewed)
idarXiv:2609.09076
affiliationYale NLP Lab
upvotes6 (Hugging Face Daily Papers, 2026-11-01)
linksabs pdf hf code
abstractAs 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 provide latent supervision for revision-oriented feedback. From real review-rebuttal threads on OpenReview, we construct ActReview-40K by aligning reviewer weaknesses with author responses and grounding the resulting feedback in localized paper evidence. We post-train Qwen3-8B-Base with multi-task supervised fine-tuning followed by GRPO using candidate-aware, weakness-specific rubric rewards. We also introduce ActReview-Bench, a human-curated benchmark of 1,000 instances for evaluating diagnostic quality and revision usefulness. Experiments show that ActReview outperforms prior specialized review-generation models on actionability and grounding while remaining competitive with strong prompt-based LLMs. Human evaluation confirms improved revision usefulness while revealing a remaining gap in technical accuracy, and additional analyses support generalization to held-out papers and robustness across independent judges.

cite

@misc{ma2026actreview,
  title={ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation},
  author={Yiling Ma and Yilun Zhao and Sihong Wu and Ziyu Chen and Manasi Patwardhan and Arman Cohan},
  year={2026},
  eprint={2609.09076},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2609.09076}
}