This article explains the key points of Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts. 2025-08-11 (arXiv) Wu, Haoyuan, Chen, Haoxing, Chen, Xiaodong, Zhou, Zhanchao, Chen, Tieyuan, Zhuang, Yihong, Lu, Guoshan, Huang, Zenan, Zhao, Junbo, Liu, Lin, et al. Inclusion AI, The Chinese University of Hong Kong,...

이번 글에서는 Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts 논문의 핵심 포인트만 간단히 정리한다. 2025년 8월 11일(Arxiv) Wu, Haoyuan, Chen, Haoxing, Chen, Xiaodong, Zhou, Zhanchao, Chen, Tieyuan, Zhuang, Yihong, Lu, Guoshan, Huang, Zenan, Zhao, Junbo, Liu, Lin, et al. Inclusion AI, The Chinese University of Hong...

Read this article in Korean Trends July 2025’s theme is how model design shapes practical reliability: the three selected papers examine how architecture and training affect reasoning visibility, local execution efficiency, and reward-based training stability. Chain of thought (CoT), the intermediate reasoning text generated before an answer or action, can...

영문판 보기 이달의 트렌드 2025년 7월의 주제는 모델 설계가 실제 사용 환경의 신뢰성에 미치는 영향이며, 선정된 3편의 논문은 아키텍처와 학습이 추론의 가시성, 로컬 실행 효율, 보상 기반 학습의 안정성을 어떻게 바꾸는지 살펴본다. 답변이나 행동 전에 생성하는 중간 추론 텍스트인 chain of thought(CoT)는 일부 유해한 의도를 드러낼 수 있다. SmallThinker는 선택적...

This article explains the key points of SmallThinker: A Family of Efficient Large Language Models Natively Trained for Local Deployment. 2025-07-28 (arXiv) Song, Yixin, Xue, Zhenliang, Wei, Dongliang, Chen, Feiyang, Gao, Jianxiang, Liu, Junchen, Liang, Hangyu, Qin, Guangshuo, Tian, Chengrong, Wen, Bo, et al. Institute of Parallel and Distributed Systems,...