Reveals that cross-lingual knowledge failure in large reasoning models is primarily a script-translation barrier rather than a linguistic or reasoning deficit.
March 19, 2026
Original Paper
Large Reasoning Models Struggle to Transfer Parametric Knowledge Across Scripts
arXiv · 2603.17070
The Takeaway
The study shows that models 'know' the information but fail to retrieve it when the query script differs from the training script. This suggests that post-training focus on transliteration and script-agnostic entity reasoning can unlock massive latent knowledge without more pre-training.
From the abstract
In this work, we analyze shortcomings in cross-lingual knowledge transfer in large, modern reasoning LLMs. We demonstrate that the perceived gap in knowledge transfer is primarily a script barrier. First, we conduct an observational data analysis on the performance of thinking models on two datasets with local knowledge from around the world, ECLeKTic and MultiLoKo. Our regression analysis shows that script match - not language or family - is the primary predictor of knowledge transfer failure o