TreeReview: A Dynamic Tree of Questions Framework for Deep and Efficient LLM-based Scientific Peer Review

TL;DR

TreeReview introduces a hierarchical, dynamic question-answering framework for LLM-based scientific peer review, improving review depth and efficiency by modeling reviews as a tree of questions and answers, with significant token reduction.

Contribution

It presents a novel hierarchical review framework with dynamic question expansion, enhancing review quality and efficiency over existing methods.

Findings

Outperforms baselines in comprehensive review generation

Reduces LLM token usage by up to 80%

Effective in generating expert-aligned feedback

Abstract

While Large Language Models (LLMs) have shown significant potential in assisting peer review, current methods often struggle to generate thorough and insightful reviews while maintaining efficiency. In this paper, we propose TreeReview, a novel framework that models paper review as a hierarchical and bidirectional question-answering process. TreeReview first constructs a tree of review questions by recursively decomposing high-level questions into fine-grained sub-questions and then resolves the question tree by iteratively aggregating answers from leaf to root to get the final review. Crucially, we incorporate a dynamic question expansion mechanism to enable deeper probing by generating follow-up questions when needed. We construct a benchmark derived from ICLR and NeurIPS venues to evaluate our method on full review generation and actionable feedback comments generation tasks. Experimental results of both LLM-based and human evaluation show that TreeReview outperforms strong baselines in providing comprehensive, in-depth, and expert-aligned review feedback, while reducing LLM token usage by up to 80% compared to computationally intensive approaches. Our code and benchmark dataset are available at https://github.com/YuanChang98/tree-review.