RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation

TL;DR

This paper introduces RbtAct, a method leveraging peer review rebuttals as implicit supervision to generate more actionable and specific review feedback, supported by a large dataset and improved training techniques.

Contribution

It proposes a new task and dataset for review feedback generation, and demonstrates a fine-tuned LLM that produces more actionable review comments.

Findings

Enhanced feedback actionability and specificity over baselines.

Built RMR-75K dataset linking review segments to rebuttals.

Model maintains relevance and grounding while improving feedback quality.

Abstract

Large language models (LLMs) are increasingly used across the scientific workflow, including to draft peer-review reports. However, many AI-generated reviews are superficial and insufficiently actionable, leaving authors without concrete, implementable guidance and motivating the gap this work addresses. We propose RbtAct, which targets actionable review feedback generation and places existing peer review rebuttal at the center of learning. Rebuttals show which reviewer comments led to concrete revisions or specific plans, and which were only defended. Building on this insight, we leverage rebuttal as implicit supervision to directly optimize a feedback generator for actionability. To support this objective, we propose a new task called perspective-conditioned segment-level review feedback generation, in which the model is required to produce a single focused comment based on the complete paper and a specified perspective such as experiments and writing. We also build a large dataset named RMR-75K that maps review segments to the rebuttal segments that address them, with perspective labels and impact categories that order author uptake. We then train the Llama-3.1-8B-Instruct model with supervised fine-tuning on review segments followed by preference optimization using rebuttal derived pairs. Experiments with human experts and LLM-as-a-judge show consistent gains in actionability and specificity over strong baselines while maintaining grounding and relevance.