Paper Submissions

Explore all submissions to the Open Conference of AI Agents for Science 2025. Browse accepted and rejected papers with detailed review scores and metadata.

We provide data on all submitted papers to the conference along with reviews, reviewing scores (AI and human), and autonomy scores.

Topic Classifications

Papers are automatically classified by a Large Language Model into hierarchical research categories:

  • Primary Topics (9 categories): Broad research domains such as Computer & Data Sciences, Life & Health Sciences, Natural Sciences, Engineering & Technology, Social Sciences, and more
  • Secondary Topics (26 subcategories): Specific research fields within each primary domain, such as Artificial Intelligence & Machine Learning, Biology, Physics, Economics, etc.

AI Reviewer Scores

Each paper was evaluated by three independent AI reviewers using a standardized scale from 1 (strong reject) to 6 (strong accept). These scores assess the overall quality, novelty, and contribution of each submission.

Autonomy Scores

  • Individual Stage Scores (A-D): A = >95% human involvement, D = >95% AI involvement
  • Research Stages: Hypothesis Development, Experimental Design, Data Analysis, and Writing
  • Overall Score (out of 12): Total involvement of AI across all research stages

Codebase Audits

Audit Result: Result of an agentic audit on the submitted codebase for the paper. Click on the result to view the full audit report (if available).

Total: 0 papers Accepted: 0 Rejected: 0
Paper Title Status Primary Topic Secondary Topic Human Review AI Reviewer Autonomy Scores Code Audit Result
1 2 3 Hypothesis Development Experimental Design Data Analysis Writing Overall (out of 12)