对齐让语言模型想法变单一，可能影响研究可靠性
One fish, two fish, but not the whole sea: Alignment reduces language models' conceptual diversity
- 用模拟人类行为的方式衡量模型概念多样性
- 对齐后的模型概念多样性比未对齐的低30%以上
- 适合关注模型公平性与多样性的研究人员
社会科学研究者近年主张用大语言模型（LLMs）替代人类进行行为研究。除了模型是否准确反映群体模式的争议外，还存在其能否捕捉人类概念多样性的疑问。同时，后训练对齐（如RLHF或RLAIF）是否影响模型内部多样性尚存争议。受人类研究启发，我们提出一种新方法：通过比较模拟个体内部变异与群体层面变异来衡量合成生成的LLM“群体”概念多样性。在两个具备丰富人类行为数据的领域中，评估了非对齐与对齐模型的表现。结果显示，无模型达到人类水平多样性，但对齐模型普遍表现出比指令微调模型更低的多样性。该发现揭示了提升模型价值对齐与降低概念表征多样性之间的潜在权衡。
原文摘要 · Abstract (English)
Researchers in social science and psychology have recently proposed using large language models (LLMs) as replacements for humans in behavioral research. In addition to arguments about whether LLMs accurately capture population-level patterns, this has raised questions about whether LLMs capture human-like conceptual diversity. Separately, it is debated whether post-training alignment (RLHF or RLAIF) affects models' internal diversity. Inspired by human studies, we use a new way of measuring the conceptual diversity of synthetically-generated LLM "populations" by relating the internal variability of simulated individuals to the population-level variability. We use this approach to evaluate non-aligned and aligned LLMs on two domains with rich human behavioral data. While no model reaches human-like diversity, aligned models generally display less diversity than their instruction fine-tuned counterparts. Our findings highlight potential trade-offs between increasing models' value alignment and decreasing the diversity of their conceptual representations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。