弥合创造力理解差距：小规模人类对齐使大语言模型达到专家级幽默排名能力
Bridging the Creativity Understanding Gap: Small-Scale Human Alignment Enables Expert-Level Humor Ranking in LLMs
AI总结：
本研究将幽默理解分解为视觉增强、推理解释和人类偏好对齐三部分，通过小规模对齐使LLM在漫画标题排名上达到82.4%准确率，超越先前67%基准并匹敌人类专家，证明聚焦对齐可有效解决创造性判断局限。
摘要：
大语言模型（LLMs）在理解创造性内容方面表现出显著局限性，Hessel等人（2023）关于《纽约客》漫画标题竞赛（NYCCC）的开创性工作证明了这一点。他们的研究揭示了LLMs与人类在幽默理解上的巨大差距，确立了理解和评估创造性内容是人工智能发展的关键挑战。我们通过将幽默理解分解为三个组成部分并系统性地改进每个部分来重新审视这一挑战：通过改进标注增强视觉理解，利用LLM生成的幽默推理和解释，以及实施针对人类偏好数据的有针对性的对齐。我们改进的方法在标题排名中达到了82.4%的准确率，显著优于之前67%的基准，并在此领域达到了世界知名人类专家的水平。值得注意的是，虽然通过不同人物角色提示模仿子群体偏好的尝试影响甚微，但使用群体偏好进行模型微调却极为有效。这些发现表明，通过针对特定子群体和个体的聚焦对齐，可以有效解决LLMs在创造性判断方面的局限性。最后，我们提出观点：实现通用人工智能需要在创造性领域系统性地收集人类偏好数据。我们主张，正如人类创造力深受个体和文化偏好影响一样，使用多样化的人类偏好数据训练LLMs可能是发展真正创造性理解的关键。
英文摘要：
Large Language Models (LLMs) have shown significant limitations in understanding creative content, as demonstrated by Hessel et al. (2023)'s influential work on the New Yorker Cartoon Caption Contest (NYCCC). Their study exposed a substantial gap between LLMs and humans in humor comprehension, establishing that understanding and evaluating creative content is key challenge in AI development. We revisit this challenge by decomposing humor understanding into three components and systematically improve each: enhancing visual understanding through improved annotation, utilizing LLM-generated humor reasoning and explanations, and implementing targeted alignment with human preference data. Our refined approach achieves 82.4% accuracy in caption ranking, singificantly improving upon the previous 67% benchmark and matching the performance of world-renowned human experts in this domain. Notably, while attempts to mimic subgroup preferences through various persona prompts showed minimal impact, model finetuning with crowd preferences proved remarkably effective. These findings reveal that LLM limitations in creative judgment can be effectively addressed through focused alignment to specific subgroups and individuals. Lastly, we propose the position that achieving artificial general intelligence necessitates systematic collection of human preference data across creative domains. We advocate that just as human creativity is deeply influenced by individual and cultural preferences, training LLMs with diverse human preference data may be essential for developing true creative understanding.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
