基于检索的长上下文翻译用于文化图像描述:佛罗里达大学Gators参加2026年美洲自然语言处理共享任务的提交
Retrieval-Augmented Long-Context Translation for Cultural Image Captioning: Gators submission for AmericasNLP 2026 shared task
AI总结:
本文提出了一种基于检索的长上下文翻译方法,用于文化图像描述,通过两阶段流程生成西班牙语中间描述,再利用检索增强的多示例提示生成目标语言描述,显著提升了Bribri、Guaraní和Orizaba Nahuatl语言的描述生成性能,并在共享任务中获得冠军。
摘要:
我们提出了佛罗里达大学Gators团队对2026年美洲自然语言处理共享任务在原住民语言文化图像描述任务中的提交。我们的两阶段流程使用Qwen2.5-VL生成西班牙语中间描述,然后利用检索增强的多示例提示与Gemini 2.5 Flash生成目标语言描述。我们在开发集评估中分别实现了Bribri、Guaraní和Orizaba Nahuatl描述生成性能的164.1%、131.7%和122.6%的提升,并在测试集评估中保持Bribri和Orizaba Nahuatl语言的>150%提升。我们发现检索高度依赖语言,仅对大规模、领域内语料有效,并且合成数据增强对开发集Guaraní性能提升贡献了约28 chrF++。我们的提交在共享任务中获得冠军,位列五份最终提交中的第二名。
英文摘要:
We present the University of Florida Gators submission to the AmericasNLP 2026 shared task on cultural image captioning for Indigenous languages. Our two-stage pipeline generates a Spanish intermediate caption with Qwen2.5-VL, then produces the target-language caption using retrieval-augmented many-shot prompting with Gemini 2.5 Flash. We achieve 164.1%, 131.7%, and 122.6% improvements over the shared task baseline for Bribri, Guaraní, and Orizaba Nahuatl captioning, respectively, in our dev set evaluation and maintain >150% improvements for the Bribri and Orizaba Nahuatl languages in the test set evaluation. We find retrieval is highly language-dependent, beneficial only for large, in-domain corpora, and that synthetic data augmentation accounts for around 28 chrF++ of the dev set Guaraní performance gain. Our submission is the overall winner of the shared task, placing second out of five finalist submissions in human evaluations of target-language captions.
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。