arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2412.12948cs.CL
收藏

MOPO:面向情感文本生成的多目标提示优化

MOPO: Multi-Objective Prompt Optimization for Affective Text Generation

Institut für Maschinelle Sprachverarbeitung, University of Stuttgart(斯图加特大学机器语言研究所);Fundamentals of Natural Language Processing, University of Bamberg(班贝格大学自然语言处理基础研究所)

Yarik Menchaca Resendiz, Roman Klinger

AI总结:

MOPO提出多目标提示优化方法,通过输出不同权重的提示集,使情感文本生成适配多领域,性能提升达15个百分点,并降低计算开销。

摘要:

情感的表达方式取决于上下文和领域。例如,在X(原推特)上,作者可能仅使用标签#anger,而在新闻标题中,情感通常以更礼貌、间接的方式书写。为了使条件文本生成模型能够创建符合领域的情感色彩文本,用户需要访问一个参数,以便选择合适的情感表达方式。为此,我们引入了MOPO,一种多目标提示优化方法。MOPO根据多个目标(此处对应于为不同领域训练的情感分类器分配的输出概率)优化提示。与单目标优化不同,MOPO输出一组提示,每个提示对多个目标具有不同的权重。用户随后可以选择最适合其上下文的提示。我们使用由各种领域特定情感分类器确定的三个目标来评估MOPO。与单目标优化相比,MOPO在所有目标上的性能提升高达15个百分点,而任何单个目标的损失最小(1-2个百分点)。这些微小的性能损失被跨多个目标的更广泛泛化所抵消——这是单目标优化无法实现的。此外,MOPO通过同时优化多个目标,减少了计算需求,无需为每个目标进行单独的优化过程。

英文摘要:

How emotions are expressed depends on the context and domain. On X (formerly Twitter), for instance, an author might simply use the hashtag #anger, while in a news headline, emotions are typically written in a more polite, indirect manner. To enable conditional text generation models to create emotionally connotated texts that fit a domain, users need to have access to a parameter that allows them to choose the appropriate way to express an emotion. To achieve this, we introduce MOPO, a Multi-Objective Prompt Optimization methodology. MOPO optimizes prompts according to multiple objectives (which correspond here to the output probabilities assigned by emotion classifiers trained for different domains). In contrast to single objective optimization, MOPO outputs a set of prompts, each with a different weighting of the multiple objectives. Users can then choose the most appropriate prompt for their context. We evaluate MOPO using three objectives, determined by various domain-specific emotion classifiers. MOPO improves performance by up to 15 pp across all objectives with a minimal loss (1-2 pp) for any single objective compared to single-objective optimization. These minor performance losses are offset by a broader generalization across multiple objectives - which is not possible with single-objective optimization. Additionally, MOPO reduces computational requirements by simultaneously optimizing for multiple objectives, eliminating separate optimization procedures for each objective.

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。

↑