What problem does it solve?
PaperJury helps researchers identify unsupported claims, weak experiments, unclear writing, and submission risks before peer review, while preventing unsafe or evidence-free revisions.
Core Features & Use Cases
- Adversarial paper review: Simulates multiple domain reviewers, routes disputed issues through a two-sided trial, and produces invalid, fixable, or author-required verdicts.
- Bounded revision workflow: Drafts minimal LaTeX or Markdown edits, requires author sign-off, tracks issues in a durable ledger, and applies risk-proportional safety checks.
- Format and submission checks: Supports LaTeX, Markdown, plain text, and one-time Word extraction, with compile validation, structural linting, compliance screening, and explicit degradation when tools are unavailable.
- Use Case: Before submitting a machine learning paper, ask PaperJury to review the experiments and claims, then use its evidence-backed ledger and verified patches to address safe issues while retaining research decisions for the author.
Quick Start
Ask PaperJury to review your paper, especially whether its experiments and claims are adequately supported.