Multi-LLM Collaborative Alignment via Stackelberg Games

Excerpt

arXiv:2609.39076v1 Announce Type: cross Abstract: A pool of language models can collaborate and improve collectively by learning from one another's responses. These interactions depend on the instructions used during training. Existing methods typically sample instructions uniformly, even though their usefulness may change as the models improve: an instruction on which models' responses once differed in quality may later be answered equally well, while a previously difficult instruction may begi