light-idea-critique

Evaluating if a research idea is novel enough requires a multi-perspective critique using cross-literature checks to verify genuine novelty. This process surfaces hidden risks and provides a structured verdict aligned with top-tier conference standards to confirm originality.

Critiquing an early-stage ML idea for feasibility involves an eight-dimension scoring framework covering theory depth, data, and execution. A five-voice critique workflow with a Devil's Advocate evaluates these dimensions to produce a revision roadmap for necessary improvements.

Getting a revision roadmap for weak theoretical proposals requires running a multi-view critique workflow that identifies weaknesses across eight dimensions. The process automatically generates a revision roadmap detailing specific improvements needed to meet top-tier conference standards.

You can batch review multiple research ideas for publishability using a reproducible evaluation pipeline. By inputting multiple idea stubs with titles and keywords, the workflow outputs structured verdicts and revision roadmaps for each idea simultaneously.

The idea critique workflow does not require existing literature references but supports optional literature validation. You only need an idea stub with a title, domain, and keywords, while adding references enhances the cross-literature novelty verification process.

You should not use an automated research idea critique when your proposal lacks a defined domain or keywords. The workflow requires an idea stub with a title, domain, and keywords to function properly and generate an accurate revision roadmap.