반응형
250x250
Notice
Recent Posts
Recent Comments
Link
| 일 | 월 | 화 | 수 | 목 | 금 | 토 |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | ||
| 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| 13 | 14 | 15 | 16 | 17 | 18 | 19 |
| 20 | 21 | 22 | 23 | 24 | 25 | 26 |
| 27 | 28 | 29 | 30 |
Tags
- dinov2 논문 리뷰
- 원격 학습 안끊기게
- 기계학습
- 이미지 필터링
- clip adapter
- clip
- evaluating object hallucination in large vision-language models paper
- evaluating object hallucination in large vision-language models
- 딥러닝 목적함수
- 논문 리뷰
- polling-based object probing evaluation
- dinov2: learning robust visual features without supervision 논문 리뷰
- vlm 환각
- 딥러닝 엔트로피
- object detection
- vlm hallucination
- 객체 검출
- 1차 미분 마스크
- vlm
- 논문 요약
- vlm hallucination paper
- dinov2: learning robust visual features without supervision
- 엔트로피란
- dinov2: learning robust visual features without supervision 논문
- vlm 환각이란
- mobilenetv1
- 에지 검출
- evaluating object hallucination in large vision-language models 논문
- Object detection article
- blip-2
Archives
- Today
- Total
목록efficientvlm: fast and accurate vision-language models via knowledge distillation and modal-adaptive pruning (1)
My Vision, Computer Vision
[논문 리뷰/요약] EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive Pruning
EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive PruningPre-trained vision-language models (VLMs) have achieved impressive results in a range of vision-language tasks. However, popular VLMs usually consist of hundreds of millions of parameters which brings challenges for fine-tuning and deployment in real-worldarxiv.org Author : Wang, Tiannan, ..
Paper
2025. 3. 31. 14:25