Quit: A Human-in-the-Loop Platform for AI Research Automation
QUIT is a human-in-the-loop platform designed to automate AI research processes while keeping researchers in control. It features a transparent pipeline that allows for the inspection and editing of intermediate outputs. The system is designed to be traceable and reproducible, with a focus on facilitating research from query to publication.
- ▪QUIT allows researchers to search, understand, implement, and draft papers through a coordinated system of specialized agents.
- ▪The platform supports an end-to-end mode, enabling users to set a topic and let the pipeline run without intervention.
- ▪Researchers can stop at any stage to review and edit outputs, ensuring full control over the research process.
Opening excerpt (first ~120 words) tap to expand
QUIT: A Human-in-the-Loop Platform for AI Research Automation Query · Understand · Implement · Tell ✦ Quit the Old Way of Doing Research ✦ QUIT is a human-in-the-loop research assistant — not a black box, but a transparent pipeline where researchers remain in full control at every step. Its artifact-driven design eliminates long-context dependency and avoids redundant token consumption. QUIT also fully supports end-to-end mode — just set a topic and let the pipeline run to completion without any intervention. End-to-end cost with DeepSeek-V4-Pro: roughly ¥10 ($1.5) per manuscript.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.