Let's Roleplay: Examining LLM Alignment in Collaborative Dialogues
This addresses the need for reliable AI collaborators in multiturn, multiparty interactions, though it is incremental as it builds on existing alignment methods with a novel evaluation framework.
The paper tackled the problem of evaluating LLM alignment in collaborative dialogues by introducing friction agents to encourage reflection, and found that a friction-aware approach significantly outperformed common baselines in improving convergence to common ground and task correctness.
As Large Language Models (LLMs) integrate into diverse workflows, they are increasingly being considered "collaborators" with humans. If such AI collaborators are to be reliable, their behavior over multiturn interactions must be predictable, validated and verified before deployment. Common alignment techniques are typically developed under simplified single-user settings and do not account for the dynamics of long-horizon multiparty interactions. This paper examines how different alignment methods affect LLM agents' effectiveness as partners in multiturn, multiparty collaborations. We study this question through the lens of friction agents that intervene in group dialogues to encourage the collaborative group to slow down and reflect upon their reasoning for deliberative decision-making. Using a roleplay methodology, we evaluate interventions from differently-trained friction agents in collaborative task conversations. We propose a novel counterfactual evaluation framework that quantifies how friction interventions change the trajectory of group collaboration and belief alignment. Our results show that a friction-aware approach significantly outperforms common alignment baselines in helping both convergence to a common ground, or agreed-upon task-relevant propositions, and correctness of task outcomes.