Google Research has developed an experimental tool using an advanced version of Gemini to provide automated feedback on theoretical computer science papers. Tested for the STOC 2026 conference, the AI tool successfully identified a range of issues, from minor typos to critical bugs in proofs, with over 80% of participants opting in. Authors reported high satisfaction, with 97% finding the feedback helpful and willing to use it again, appreciating its speed and rigorous, neutral tone. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
RANK_REASON This describes an experimental tool using an AI model to assist in the peer-review process for academic papers, which falls under research and product development.
