Paper Copilot tracks ICML 2026 community scores

Paper Copilot has become the "shadow leaderboard" for top-tier AI conferences, filling the transparency gap left by official platforms like OpenReview.

• –Real-time data logging shows a score range of 1.0 to 6.0, helping authors benchmark their "borderline" ratings against the community.
• –Historical context from ICML 2024/2025 (averaging 3.23 for accepted papers) helps manage expectations and strategy during the rebuttal phase.
• –Area-specific filtering reveals disparate grading curves across sub-disciplines like Deep Learning and Reinforcement Learning.
• –The platform's popularity underscores the community's demand for data-driven insights into the increasingly competitive and opaque peer-review process.