ReciLoReL 基准发布:覆盖豪萨语、约鲁巴语、伊博语的菜谱翻译评测
Cross-cultural Adaptation of Recipes for Low-Resource Languages: A Benchmark and Multilingual Culinary Lexicon for Hausa, Yorùbá, and Igbo
尼日利亚三种语言的菜谱翻译基准加上 1.5 万词条的烹饪词典,做低资源语言本地化的团队可以直接拿来用,不用微调也能靠词典提升翻译质量。
研究团队发布 ReciLoReL 基准,包含 600 篇经专家翻译并做文化适配的菜谱,覆盖尼日利亚三大语言 Yorùbá、Hausa、Igbo,同时推出约 1.5 万条翻译实体的多语言烹饪词典 ReciLex,涵盖食材、厨具、厨师动作、工具状态、时长等十个类别。
菜谱翻译的难点在于不只是语言转换,还需要把食材、计量单位和烹饪流程适配到目标文化语境。评测方法除 BLEU、chrF、TER 三项常规指标外,另引入基于跨文化适应方法的五项文化适配评估标准,采用五分制人工评分。将 ReciLex 接入多语言 LLM 后,无需微调,平均 BLEU 从 13.57 提升至 14.51,chrF 从 37.89 提升至 38.74,TER 从 77.14 降至 75.12,人工评估的五项文化适配维度提升 0.11 至 0.36 分。
这项工作填补了低资源语言文化适配翻译的资源空白,为没有大规模目标语料训练条件的研究者和本地化团队提供了词典增强、零样本/少样本的可行路径。
Cross-cultural Adaptation of Recipes for Low-Resource Languages: A Benchmark and Multilingual Culinary Lexicon for Hausa, Yorùbá, and Igbo
Recent multilingual large language models (LLMs) have improved machine translation, yet culturally adaptive translation for low-resource languages (LRLs) remains underexplored. This is particularly challenging for recipes, where ingredients, cooking tools, measurements, and procedures may require adaptation to the target cultural context. In this work, we introduce ReciLoReL, a benchmark of 600 expert-translated and culturally adapted recipes across Yorùbá, Hausa, and Igbo (three of the most prominent languages in Nigeria), and ReciLex, a multilingual culinary lexicon containing approximately 15,000 translated entities across ten recipe entity categories, including ingredients, tools, chef actions, state of tools, and durations. We evaluate state-of-the-art multilingual LLMs for English-to-LRL recipe translation under zero-shot and few-shot settings and investigate whether augmenting LLMs with ReciLex improves culturally adaptive recipe translation without requiring large target-language recipe corpora for training or fine-tuning. Beyond conventional reference-based translation evaluation using BLEU, chrF, and TER, we introduce five criteria for evaluating cultural adaptation, adapted from established cross-cultural adaptation methods. Across the selected LLMs and languages, ReciLex improves average BLEU from 13.57 to 14.51 and chrF from 37.89 to 38.74, while reducing TER from 77.14 to 75.12. Human evaluation further shows improvements ranging from 0.11 to 0.36 points across the five cultural adaptation evaluation criteria on a five-point scale. Overall, our work provides new resources, evaluation criteria, and empirical evaluation for studying culturally adaptive recipe translation in low-resource settings.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include ReciLoReL, low-resource languages, Hausa, Yorùbá.
Source attribution: curated from arXiv cs.AI (@Nuhu Ibrahim, Maryam Abdulazeez, Fatsuma Jauro, Salisu Aliyu, Robert Stevens, Riza Batista-Navarro).