The task of persona-steered text generation requires large language models
(LLMs) to generate text that reflects the distribution of views that an
individual fitting a persona could have. People have multifaceted personas, but
prior work on bias in LLM-generated opinions has only explored multiple-choice
settings or one-dimensional personas. We define an incongruous persona as a
persona with multiple traits where one trait makes its other traits less likely
in human survey data, e.g. political liberals who support increased military
spending. We find that LLMs are 9.7% less steerable towards incongruous
personas than congruous ones, sometimes generating the stereotypical stance
associated with its demographic rather than the target stance. Models that we
evaluate that are fine-tuned with Reinforcement Learning from Human Feedback
(RLHF) are more steerable, especially towards stances associated with political
liberals and women, but present significantly less diverse views of personas.
We also find variance in LLM steerability that cannot be predicted from
multiple-choice opinion evaluation. Our results show the importance of
evaluating models in open-ended text generation, as it can surface new LLM
opinion biases. Moreover, such a setup can shed light on our ability to steer
models toward a richer and more diverse range of viewpoints.

存在困扰个性化驱动文本生成任务的大型语言模型（LLMs）需要生成反映符合特定人物角色可能具有的观点分布的文本。我们定义不协调的人物角色作为具有多个特征的人物角色，在人类调查数据中，其中一个特征会使其其他特征变得不太可能，例如支持增加军费的政治自由主义者。我们发现 LLMs 比一致的人物角色少 9.7％的可控性，有时会生成与其人口统计学相关的刻板立场，而不是目标立场。我们评估的使用人类反馈强化学习（RLHF）微调的模型更具可控性，尤其是与政治自由主义者和女性相关的立场，但呈现出更少多样化的人物角色观点。我们还发现无法从多选意见评估中预测的 LLM 可控性的方差。我们的研究结果显示了在开放式文本生成中评估模型的重要性，因为它可以揭示新的 LLM 观点偏见。此外，这种设置可以揭示我们将模型引导到更丰富和多样化的观点范围的能力。