Search papers, labs, and topics across Lattice.
This paper explores methods for generating socially intelligent compromises between opposing viewpoints using LLMs, focusing on place-based conflicts. They compared four prompt engineering methods with Claude 3 Opus, finding that iterative feedback based on external empathic similarity between the compromise and each viewpoint yields the most acceptable compromises. They then distilled this approach into smaller models via margin-based alignment, improving efficiency and removing the need for empathy estimation at inference time.
LLMs can learn to generate better compromises by iteratively incorporating feedback on how empathically similar a compromise is to each viewpoint, opening the door to more socially intelligent AI.
Large Language Models (LLMs) excel academically but struggle with social intelligence tasks, such as creating good compromises. In this paper, we present methods for generating empathically neutral compromises between two opposing viewpoints. We first compared four different prompt engineering methods using Claude 3 Opus and a dataset of 2,400 contrasting views on shared places. A subset of the gen erated compromises was evaluated for acceptability in a 50-participant study. We found that the best method for generating compromises between two views used external empathic similarity between a compromise and each viewpoint as iterative feedback, outperforming stan dard Chain of Thought (CoT) reasoning. The results indicate that the use of empathic neutrality improves the acceptability of compromises. The dataset of generated compromises was then used to train two smaller foundation models via margin-based alignment of human preferences, improving efficiency and removing the need for empathy estimation during inference.