It was interesting to reproduce one of the ICML accepted papers using these tools.
This analysis from the hackathon was particularly insightful, and agent-based reproducibility has the potential to become a de facto standard for strengthening the credibility of claims in ML conferences. It could also serve as a valuable pathway toward refining autonomous scientific discovery, revealing not only where agents succeed, but also where they fail, providing important insights into their limitations and how human-agent collaboration can be improved.
🔄 In a Training Loop
Tejasri N
tejasri19
AI & ML interests
AI Researcher | AI & Science
Recent Activity
commentedon an article 7 days ago
What We Learned by Reproducing 2,200 papers from ICML upvoted an article 7 days ago
What We Learned by Reproducing 2,200 papers from ICML updated a Space 30 days ago
tejasri19/repro-mindflow-mind-supernet-powered-thinking-flows-for-research-idea-innovationOrganizations
commented on What We Learned by Reproducing 2,200 papers from ICML 7 days ago
upvoted an article 7 days ago
Article
What We Learned by Reproducing 2,200 papers from ICML
abidlabs
• • 95upvoted a paper 4 months ago
upvoted a collection about 1 year ago
upvoted an article over 1 year ago
Article
Reduce, Reuse, Recycle: Why Open Source is a Win for Sustainability
sasha
• • 17