[논문리뷰] Paying Less Generalization Tax: A Cross-Domain Generalization Study of RL Training for LLM AgentsarXiv에 게시된 'Paying Less Generalization Tax: A Cross-Domain Generalization Study of RL Training for LLM Agents' 논문에 대한 자세한 리뷰입니다.#Review#LLM Agents#Reinforcement Learning#Cross-Domain Generalization#State Information Richness#Planning Complexity#State Augmentation#Step-by-Step Reasoning#Mid-Training2026년 1월 26일댓글 수 로딩 중