CultureConverse is introduced, a scalable, multilingual simulation and evaluation harness for culturally grounded assistant dialogue that covers 10 East and Southeast Asian regions, 58 subgroup identities, and 7 domains and performance gains from fine-tuning on 27,860 high-quality CultureConverse-DS samples improve in-...
Bryan Chen Zhengyu Tan, Wei-Hua Zheng, Thong T. Doan et al.· 0 citations
BaCVA is proposed, an inference-time Bayesian Context-aware personalized Value Alignment method that approximates posterior personalized preferences by integrating static personal values with scenario-specific value salience and first estimates contextual value salience from generally normative responses.
Han-Ze Guo, Ai-Xuan Song, Jing Yao et al.· 0 citations
A unified framework for evaluating LLM uncertainty based on Hill numbers, a family of diversity measures, is introduced, which provides a common and intuitive scale for interpreting variation in LLM outputs while accommodating different notions of semantic similarity and different sensitivities to output distributions.
The causes of modal divergence are probed, offering insights into fostering culturally robust MLLMs, and a Multilingual, Multimodal Alignment framework for Cultural grounding evaluation is proposed.
Weihua Zheng, Zhengyuan Liu, Tanmoy Chakraborty et al.· Annual Meeting of the Associ...· 0 citations
CultureConverse is introduced, a scalable, multilingual simulation and evaluation harness for culturally grounded assistant dialogue that covers 10 East and Southeast Asian regions, 58 subgroup identities, and 7 domains and performance gains from fine-tuning on 27,860 high-quality CultureConverse-DS samples improve in-...
Bryan Chen Zhengyu Tan, Weihua Zheng, Thong T. Doan et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
