Tag
Jenova AI's Tour Guide agent turns photos of landmarks, menus, or street scenes into cultural stories and context-aware travel guidance, including etiquette, timing, and nearby recommendations within a 10-minute walk.
CuBEs 提出一种文化情境化的 LLM 行为评估框架，通过在测试场景中注入文化语境并在 12 种文化上构建人工标注数据集，评估了 13 个开源与闭源模型，发现文化盲的评估会遗漏非西方特有的偏见轴向（如宗教与殖民议题）。
The paper introduces MemeCULT-1K, a multilingual benchmark of 1,000 South Asian memes to evaluate vision-language models' understanding of cultural context and humor, demonstrating that providing cultural context improves model performance.
Introduces Inspect India Evals, an open-source framework for evaluating LLMs in Indian linguistic and cultural contexts, with six benchmarks testing multilingual ability, bias, safety, and cultural knowledge. Tests on five models show Sarvam-M 24B and Gemma 2 27B lead.
This paper introduces Ex-ToxiCN-MM, the first Chinese harmful meme explanation dataset, along with a knowledge base C-HarmKB and an attribution analysis framework RIKE, to improve interpretable detection of harmful memes by considering cultural context and ambiguity.