Bingni Zhang

Token Alignment Heads: Unveiling Attention's Role in LLM Multilingual Translation

LayerMix Law: Scaling Law for Large Language Models on Quality-Weighted Mixture Data with Repetition

MuBench: Assessment of Multilingual Capabilities of Large Language Models Across 61 Languages

From Translation to Multilinguality: Revisit the Role of Parallel Data in Multilingual LLM Pretraining

ParaRater: Enhancing Cross-Lingual Transfer in Large Langauge Models with Meta-Learning

QUADMIX: QUALITY-DIVERSITY BALANCED DATA SELECTION FOR EFFICIENT LLM PRETRAINING

MuRating: A High Quality Data Selecting Approach to Multilingual Large Language Model Pretraining

Exploring Polyglot Harmony: On Multilingual Data Allocation for Large Language Models Pretraining

Frame-Voyager: Learning to Query Frames for Video Large Language Models