Boundary Suppression Asymmetry in Post-trained Assistants: Over-expansion as a Controllability Cost
Researchers have identified a phenomenon called boundary suppression asymmetry in post-trained language model assistants. This asymmetry means that while these assistants are trained to be helpful and complete, it becomes harder to suppress certain helpful tendencies, like over-answering or providing too much information, when explicitly asked for narrower responses. The study suggests this is due to a combination of content budget overshoot and continuation persistence, making boundary correction more difficult for specific helpful assistant behaviors. AI
IMPACT Highlights potential challenges in fine-tuning AI assistants for precise control over response length and detail.