AI RESEARCH
PostTrainBench: Can LLM Agents Automate LLM Post-Training?
arXiv CS.LG
•
ArXi:2603.08640v1 Announce Type: cross AI agents have become surprisingly proficient at software engineering over the past year, largely due to improvements in reasoning capabilities. This raises a deeper question: can these systems extend their capabilities to automate AI research itself? In this paper, we explore post-