OpenNovelty: An LLM-powered Agentic System for Verifiable Scholarly Novelty Assessment

TL;DR

OpenNovelty introduces an LLM-powered agentic system for transparent, evidence-based scholarly novelty assessment grounded in retrieved real papers. The framework unfolds in four phases—information extraction, retrieval, analysis/synthesis, and report generation—to produce verifiable novelty judgments with explicit citations and evidence snippets. It is deployed on 500+ ICLR 2026 submissions, demonstrating the ability to identify relevant prior work that authors may overlook and offering a scalable tool for fair, consistent peer review. Future work includes systematic evaluation via NoveltyBench and broader end-to-end assessments across submissions.

Abstract

Evaluating novelty is critical yet challenging in peer review, as reviewers must assess submissions against a vast, rapidly evolving literature. This report presents OpenNovelty, an LLM-powered agentic system for transparent, evidence-based novelty analysis. The system operates through four phases: (1) extracting the core task and contribution claims to generate retrieval queries; (2) retrieving relevant prior work based on extracted queries via semantic search engine; (3) constructing a hierarchical taxonomy of core-task-related work and performing contribution-level full-text comparisons against each contribution; and (4) synthesizing all analyses into a structured novelty report with explicit citations and evidence snippets. Unlike naive LLM-based approaches, \textsc{OpenNovelty} grounds all assessments in retrieved real papers, ensuring verifiable judgments. We deploy our system on 500+ ICLR 2026 submissions with all reports publicly available on our website, and preliminary analysis suggests it can identify relevant prior work, including closely related papers that authors may overlook. OpenNovelty aims to empower the research community with a scalable tool that promotes fair, consistent, and evidence-backed peer review.

Figures (1)

• Figure 1: Overview of the OpenNovelty framework. Phase I extracts the core task and claimed contributions and generates expanded queries. Phase II retrieves and filters candidate prior work. Phase III constructs a taxonomy and performs evidence-verified comparisons. Phase IV renders the final novelty report from structured outputs.