LLM(21)
-
[논문리뷰] Selective Reflection-Tuning: Student-Selected Data Recycling for LLM Instruction-Tuning
Link: https://aclanthology.org/2024.findings-acl.958/ ACL2024 findings에 accept된 paper이다. Teacher-student model.. 모델의 distillation에서 자주 보던 용어다. 여기서도 같은 의미로 사용되는데 모델의 경량화와 함께 따라오는 학습 속도의 개선, 그러면서도 성능 유지를 위해서 이러한 방식을 채택한다. 이 논문에서는 data의 효율성을 위해 새로운 데이터의 수집 없이도 student 모델의 성능향상이 가능하다고 주장한다.내용이 엄청 쉽지는 않았어서 나름 이해하기 쉽게 작성해봤습니다.Distillation은 잘 모르는 분야기도 하고, 열심히 이해해봤는데 틀린 부분이 있을 수 있어요..ㅠㅠAbstractProblem: ..
2024.10.15 -
[논문리뷰] LLM-Adapters: An Adapter Family for Parameter-Efficient Fine-tuning of Large Language Models
Zhiqiang Hu, Lei Wang, Yihuai Lan, Wanyu Xu, Ee-Peng Lim, Lidong Bing, Xing Xu, Soujanya Poria, and Roy Lee. 2023. LLM-Adapters: An Adapter Family for Parameter-Efficient Fine-Tuning of Large Language Models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 5254–5276, Singapore. Association for Computational Linguistics. https://arxiv.org/abs/2304...
2024.10.14 -
[논문리뷰] Open Ko-LLM Leaderboard: Evaluating Large Language Models in Korean with Ko-H5 Benchmark
Chanjun Park, Hyeonwoo Kim, Dahyun Kim, SeongHwan Cho, Sanghoon Kim, Sukyung Lee, Yungi Kim, and Hwalsuk Lee. 2024. Open Ko-LLM Leaderboard: Evaluating Large Language Models in Korean with Ko-H5 Benchmark. In Proceedings of the 62nd Annual Meeting of the Association for Computational Liguistics (Volume 1: Long Papers), pages 3220-3234, Bangkok, Thailand. Association for Computational Linguistics..
2024.10.07 -
[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..
2024.09.30 -
[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..
2024.09.27 -
[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 ..
2024.09.26