목록분류 전체보기 (86)

My Vision, Computer Vision

[논문 요약/리뷰] DoRA: Weight-Decomposed Low-Rank Adaptation

DoRA: Weight-Decomposed Low-Rank AdaptationAmong the widely used parameter-efficient fine-tuning (PEFT) methods, LoRA and its variants have gained considerable popularity because of avoiding additional inference costs. However, there still often exists an accuracy gap between these methods and fullarxiv.org Author : Liu, Shih-Yang, et al.Journal : ICML 2024Keyword : DoRAPublished Date : 2024년 2월..
Paper 2025. 3. 18. 16:46
[논문 요약/리뷰] Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate ShiftTraining Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful paramarxiv.org Author : Ioffe, Sergey, and Christian Sz..
Paper 2025. 3. 17. 20:14
[논문 요약/리뷰] LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS

LoRA: Low-Rank Adaptation of Large Language ModelsAn important paradigm of natural language processing consists of large-scale pre-training on general domain data and adaptation to particular tasks or domains. As we pre-train larger models, full fine-tuning, which retrains all model parameters, becomes learxiv.orgJournal : ICLR 2022Published Date : 2021년 6월 17일Keyword : LLM, RANK Abstract모델의 크기가..
Paper 2025. 3. 11. 19:54
[논문 요약/리뷰] FuseMix: Data-Efficient Multimodal Fusion on a Single GPU

Data-Efficient Multimodal Fusion on a Single GPUThe goal of multimodal alignment is to learn a single latent space that is shared between multimodal inputs. The most powerful models in this space have been trained using massive datasets of paired inputs and large-scale computational resources, making tharxiv.orgJournal: CVPR 20204Published Date: 2023년 12월 15일Keyword: Single GPU, Vision Language ..
Paper 2025. 3. 7. 16:41
[초간단] Ubuntu 디렉터리 용량 확인 명령어 모음

1. 전체 디스크 사용량 및 마운트된 디스크별 사용량 확인# 시스템 전체 디스크 용량, 사용량 및 사용 가능한 공간 확인df -h2. 현재 디렉터리 내 폴더/파일 용량 확인 (기본적인 확인)du -hs * 3. 현재 디렉터리 내 폴더별 용량 정렬# 현재 디렉터리에 있는 모든 폴더를 크기 순으로 출력du -h --max-depth=1 . | sort -hrdu -h: 사람이 읽기 쉬운(human-readable) 단위(K, M, G)로 표시--max-depth=1: 현재 디렉터리의 1단계 하위 폴더까지만 출력sort -hr: 용량 기준으로 내림차순 정렬 4. 특정 폴더 내에서 가장 큰 10개 폴더 찾기# 홈 디렉터리(~/) 내에서 용량이 가장 큰 10개 폴더 출력du -h ~/ | sort -hr | he..
환경 설정 2025. 3. 6. 10:43
[논문 리뷰/요약] TinyLLaVA: A Framework of Small-scale Large Multimodal Models

TinyLLaVA: A Framework of Small-scale Large Multimodal ModelsWe present the TinyLLaVA framework that provides a unified perspective in designing and analyzing the small-scale Large Multimodal Models (LMMs). We empirically study the effects of different vision encoders, connection modules, language models, training darxiv.orgJournal: ArxivPublished Date: 2024년 2월 22일본 논문은 TinyLLaVA 프레임워크를 소개한다.또한..
Paper 2025. 3. 4. 19:12
이전 Prev 1 2 3 4 5 6 7 ··· 15 Next 다음