이번 글에서는 SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training 논문의 핵심 포인트만 간단히 정리한다. 2025년 1월 28일(Arxiv), Proceedings of the 42nd International Conference on Machine Learning, PMLR 267, 2025 Chu, Tianzhe, Zhai, Yuexiang, Yang, Jihan, Tong, Shengbang, Xie, Saining, Schuurmans, Dale, Le, Quoc V., Levine, Sergey,...

This article explains the key points of Open Problems in Mechanistic Interpretability. 2025-01-27 (arXiv) Sharkey, Lee, Chughtai, Bilal, Batson, Joshua, Lindsey, Jack, Wu, Jeff, Bushnaq, Lucius, Goldowsky-Dill, Nicholas, Heimersheim, Stefan, Ortega, Alejandro, Bloom, Joseph, et al. Apollo Research, Anthropic, Decode Research, Eleuther AI, FAR AI, Google DeepMind, Leap Laboratories, Harvard...

이번 글에서는 Open Problems in Mechanistic Interpretability 논문의 핵심 포인트만 간단히 정리한다. 2025년 1월 27일(Arxiv) Sharkey, Lee, Chughtai, Bilal, Batson, Joshua, Lindsey, Jack, Wu, Jeff, Bushnaq, Lucius, Goldowsky-Dill, Nicholas, Heimersheim, Stefan, Ortega, Alejandro, Bloom, Joseph, et al. Apollo Research, Anthropic, Decode Research, Eleuther AI, FAR AI, Google DeepMind, Leap Laboratories,...

이번 글에서는 AdaptLLM: Adapting-large-language-models-to-domains-via-reading-comprehension 논문의 핵심 포인트만 간단히 정리한다. 2023년 7월(Arxiv), ICLR 2024 Daixuan Cheng, Shaohan Huang B & Furu Wei. Microsoft Research,Beijing Institute for General Artificial Intelligence (BIGAI) 논문 링크 Github Huggingface 요약 Domain-specific corpora에서 pre-training 하는 것은 모델이 해당 domain에 대한 지식을 학습할 수 있도록 해 주지만, 반대로...

This article explains the key points of Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps. 2025-01-16 (arXiv) Ma, Nanye, Tong, Shangyuan, Jia, Haolin, Hu, Hexiang, Su, Yu-Chuan, Zhang, Mingda, Yang, Xuan, Li, Yandong, Jaakkola, Tommi, Jia, Xuhui, et al. NYU, MIT, Google Paper Read this article in Korean Summary...