본문 바로가기
728x90

Paper Review11

[논문리뷰] What Language Model Architecture and Pretraining Objective Work Best for Zero-Shot Generalization? What Language Model Architecture and Pretraining Objectvie Work Best for Zero-Shot Generalization, International Conference on Machine Learning, PMLR (Proceedings of Machine Learning Research), 2022.https://arxiv.org/abs/2204.05832 What Language Model Architecture and Pretraining Objective Work Best for Zero-Shot Generalization?Large pretrained Transformer language models have been shown to exhi.. 2024. 9. 30.
[논문리뷰] AI models collapse when trained on recursively generated data Shumailov, I., Shumaylov, Z., Zhao, Y. et al. AI models collapse when trained on recursively generated data. Nature 631, 755–759 (2024). https://doi.org/10.1038/s41586-024-07566-y 2024년 7월 Nature 저널에 등록된 article이다. 요즘 LLM에서 대부분의 논문들이 model의 성능을 높이거나 효율성을 향상시키는 방향이라는 점과 대비되는 논문인데, 이 논문은 생성형 AI가 만든 데이터를 LLM에 다시 학습을 시켰을 때 일어날 수 있는 model collapse 현상에 대해서 다루는 논문이라 신선했다. AbstractStable diffusion은 imag.. 2024. 9. 27.
[논문리뷰] Direct Preference Optimization: Your Language Model is Secretly a Reward Model (DPO) Direct Preference Optimization: Your Language Model is Secretly a Reward Model, Advances in Neural Information Processing Systems (Neurips,'24). Rafailov, R., Sharma, A., Mitchell, E., Manning, C. D., Ermon, S., & Finn, C.https://arxiv.org/abs/2305.18290 Direct Preference Optimization: Your Language Model is Secretly a Reward ModelWhile large-scale unsupervised language models (LMs) learn broad .. 2024. 9. 26.
[논문리뷰] LIMA: Less Is More for Alignment LIMA: Less Is More for Alignment, In Proceedings of the 37th International Conference on Neural Information Processing Systems (NIPS '23).  Chunting Zhou, Pengfei Liu, Puxin Xu, Srini Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, Lili Yu, Susan Zhang, Gargi Ghosh, Mike Lewis, Luke Zettlemoyer, and Omer Levy.https://arxiv.org/abs/2305.11206 LIMA: Less Is More for AlignmentLarge lang.. 2024. 9. 23.
[논문리뷰] PARDEN, Can You Repeat That? Defending against Jailbreaks via Repetition PARDEN, Can You Repeat That? Defending against Jailbreaks via Repetition, Accepted at ICML 2024, Ziyang Zhang, Qizhen Zhang, Jakob Foerster https://arxiv.org/abs/2405.07932 PARDEN, Can You Repeat That? Defending against Jailbreaks via RepetitionLarge language models (LLMs) have shown success in many natural language processing tasks. Despite rigorous safety alignment processes, supposedly safety.. 2024. 9. 14.
728x90

티스토리툴바