arXiv Artificial Intelligence

A-Evolve-Training: Autonomous Post-Training of a 30B Model

9 hours ago 10.09.2026 07:00 👁 2 views
A-Evolve-Training: Autonomous Post-Training of a 30B Model

Quick summary

arXiv:2606.20657v3 Announce Type: replace Abstract: Post-training a frontier model is normally weeks of human work: proposing data and recipe changes, launching runs, reading evals, deciding what to keep. We report an autonomous system that runs this loop with no human in the loop, post-training a 30B Nemotron across four rounds over multiple weeks. The autonomously produced model reaches a held-out score of 0.86 against the top human submission's 0.87 on the public NVIDIA Nemotron-Reasoning Challenge leaderboard, placing 8th of ~4000 at the time of writing. More striking than the number: the

Key takeaways

  • arXiv:2606.20657v3 Announce Type: replace Abstract: Post-training a frontier model is normally weeks of human work: proposing data and recipe changes, launching runs, reading evals, deciding what to keep.
  • We report an autonomous system that runs this loop with no human in the loop, post-training a 30B Nemotron across four rounds over multiple weeks.
  • The autonomously produced model reaches a held-out score of 0.86 against the top human submission's 0.87 on the public NVIDIA Nemotron-Reasoning Challenge leaderboard, placing 8th of ~4000 at the time of writing.

Why it matters

“A-Evolve-Training: Autonomous Post-Training of a 30B Model” should be evaluated beyond branding and benchmark scores. Its practical importance will emerge in task accuracy, latency, unit cost, safety and integration with real workflows.

Kaynak sitede devamını oku: arXiv Artificial Intelligence ↗