MARG: Multi-Agent Review Generation for Scientific Papers

Mike D'Arcy, Tom Hope, Larry Birnbaum, Doug Downey
1/8/2024
cs.CL

No official code found. Best implementation: allenai/marg-reviewer (64 stars, Apache-2.0, last commit 2026-03-05).

Abstract

We study the ability of LLMs to generate feedback for scientific papers and develop MARG, a feedback generation approach using multiple LLM instances that engage in internal discussion. By distributing paper text across agents, MARG can consume the full text of papers beyond the input length limitations of the base LLM, and by specializing agents and incorporating sub-tasks tailored to different comment types (experiments, clarity, impact) it improves the helpfulness and specificity of feedback. In a user study, baseline methods using GPT-4 were rated as producing generic or very generic comments more than half the time, and only 1.7 comments per paper were rated as good overall in the best baseline. Our system substantially improves the ability of GPT-4 to generate specific and helpful feedback, reducing the rate of generic comments from 60% to 29% and generating 3.7 good comments per paper (a 2.2x improvement).

View on arXivView PDF

Code Implementations(1)

allenai/marg-reviewerCommunity100%
Apache-2.0

Code/data for MARG (multi-agent review generation)

649CSS, HTMLDec 20, 20237 months agoApache-2.0

Similar Papers

ReviewAgents: Bridging the Gap Between Human and AI-Generated Paper Reviews

84% match2025

Generative Adversarial Reviews: When LLMs Become the Critic

83% match2024

FRAME: Feedback-Refined Agent Methodology for Enhancing Medical Research Insights

82% match2025

LLM Review: Enhancing Creative Writing via Blind Peer Review Feedback

82% match2026

Unveiling the Merits and Defects of LLMs in Automatic Review Generation for Scientific Papers

81% match1 repo32025

MARS: toward more efficient multi-agent collaboration for LLM reasoning

81% match1 repo62025

Cite this paper

@article{darcy2024marg,
  title  = {MARG: Multi-Agent Review Generation for Scientific Papers},
  author = {Mike D'Arcy and Tom Hope and Larry Birnbaum and Doug Downey},
  year   = {2024},
  eprint = {2401.04259},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2401.04259v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2401.04259.svg)](https://codeofpaper.com/paper/2401.04259)

Discussion