Source-Grounded Semantic Reinforcement Learning for Low-Resource Target-Language Generation 事件

Source-Grounded Semantic Reinforcement Learning for Low-Resource Target-Language Generation arXiv:2605.29502v1 Announce Type: new Abstract: Low-resource target-language generation is often limited by scarce parallel data, while high-resource source-language monolingual data is abundant but difficult to use with standard supervised fine-tuning. We propose Source-Grounded Semantic Reinforcement Learning (SG-SRL), a resource-utilization framework that converts source-language monolingual data into