PostTrainBench

'PostTrainBench' is a benchmark dataset designed to evaluate the performance of AI agents in curating high-quality training datasets for the post-training phase of aligning large language models (LLMs).

Papers using PostTrainBench (2)

AutoTrainess: Teaching Language Models to Improve Language Models Autonomously2026ANDES: Agent Native Data Evolving Synthesis Tool for Autonomous Instruction Alignment2026