arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.07993cs.CLcs.AI
收藏

利用作者特定上下文进行科学图表标题生成:第三届SciCap挑战赛

Leveraging Author-Specific Context for Scientific Figure Caption Generation: 3rd SciCap Challenge

Department of Biomedical Engineering, Faculty of Engineering, Mahidol University(生物医学工程系,工程学院,玛希多大学);Faculty of Medicine Ramathibodi Hospital, Mahidol University(拉玛蒂波迪医院医学院,玛希多大学)

Watcharapong Timklaypachara, Monrada Chiewhawan, Nopporn Lekuthai, Titipat Achakulvisut

AI总结:

针对第三届SciCap挑战赛,提出两阶段流水线,结合上下文过滤、类别特定提示优化和作者风格细化,生成科学准确且风格一致的图表标题,显著提升ROUGE和BLEU分数。

摘要:

科学图表标题需要兼具准确性和风格一致性,以传达视觉信息。在此,我们为第三届SciCap挑战赛提出了一种领域特定的标题生成系统,该系统利用LaMP-Cap数据集,将图表相关的文本上下文与作者特定的写作风格相结合。我们的方法采用两阶段流水线:第一阶段结合上下文过滤、通过DSPy的MIPROv2和SIMBA进行的类别特定提示优化,以及标题候选选择;第二阶段应用带配置文件图的少样本提示进行风格细化。我们的实验表明,类别特定提示优于零样本和通用优化方法,将ROUGE-1召回率提高了+8.3%,同时将精确率损失限制在-2.8%,BLEU-4降低限制在-10.9%。基于配置文件的风格细化在BLEU分数上取得了40-48%的提升,在ROUGE上取得了25-27%的提升。总体而言,我们的系统表明,将上下文理解与作者特定的风格适应相结合,可以生成既科学准确又忠实于源论文风格的标题。

英文摘要:

Scientific figure captions require both accuracy and stylistic consistency to convey visual information. Here, we present a domain-specific caption generation system for the 3rd SciCap Challenge that integrates figure-related textual context with author-specific writing styles using the LaMP-Cap dataset. Our approach uses a two-stage pipeline: Stage 1 combines context filtering, category-specific prompt optimization via DSPy's MIPROv2 and SIMBA, and caption candidate selection; Stage 2 applies few-shot prompting with profile figures for stylistic refinement. Our experiments demonstrate that category-specific prompts outperform both zero-shot and general optimized approaches, improving ROUGE-1 recall by +8.3\% while limiting precision loss to -2.8\% and BLEU-4 reduction to -10.9\%. Profile-informed stylistic refinement yields 40--48\% gains in BLEU scores and 25--27\% in ROUGE. Overall, our system demonstrates that combining contextual understanding with author-specific stylistic adaptation can generate captions that are both scientifically accurate and stylistically faithful to the source paper.

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。

↑