DeepReviewer 2.0: A Traceable Agentic System for Auditable Scientific Peer Review

TL;DR

DeepReviewer 2.0 is an advanced, traceable AI system designed for transparent and auditable scientific peer reviews, outperforming some existing models in coverage and comparison tests.

Contribution

It introduces a process-controlled, traceable review system that produces audit-friendly critique packages with anchored evidence and follow-up actions.

Findings

Outperforms Gemini-3.1-Pro in coverage and blind comparison tests.

Achieves 71.63% win rate against human reviewers in blind evaluations.

Builds a claim-evidence-risk ledger and verification agenda for transparent reviews.

Abstract

Automated peer review is often framed as generating fluent critique, yet reviewers and area chairs need judgments they can \emph{audit}: where a concern applies, what evidence supports it, and what concrete follow-up is required. DeepReviewer~2.0 is a process-controlled agentic review system built around an output contract: it produces a \textbf{traceable review package} with anchored annotations, localized evidence, and executable follow-up actions, and it exports only after meeting minimum traceability and coverage budgets. Concretely, it first builds a manuscript-only claim--evidence--risk ledger and verification agenda, then performs agenda-driven retrieval and writes anchored critiques under an export gate. On 134 ICLR~2025 submissions under three fixed protocols, an \emph{un-finetuned 196B} model running DeepReviewer~2.0 outperforms Gemini-3.1-Pro-preview, improving strict major-issue coverage (37.26\% vs.\ 23.57\%) and winning 71.63\% of micro-averaged blind comparisons against a human review committee, while ranking first among automatic systems in our pool. We position DeepReviewer~2.0 as an assistive tool rather than a decision proxy, and note remaining gaps such as ethics-sensitive checks.