Mohammad Atif Quamar

Multi-agent collaboration bottlenecks when agent personas stay rigid, but dynamic, test-time strategy transplantation across system prompts enables heterogeneous LLMs to collectively align on the most effective reasoning path.

OPSD gains stem more from the teacher's contextual influence than from privileged access to problem-specific solutions.