AGENTTUNING

Summary from the TLDR AI newsletter

Instruction-tune LLMs using interaction trajectories across multiple agent tasks.