-
- https://arxiv.org/abs/2103.08849
- NAACL2021
本論文では、視覚言語モデルのゼロショットの言語間転送について研究している。
具体的には、多言語テキスト-ビデオ検索に焦点を当て、文脈に応じた多言語マルチモーダル埋め込みを学習するTransformerベースのモデルを提案する。
ゼロショットの設定では、英語以外の文章を多言語テキスト-ビデオモデ…
e4exp updated
3 years ago
-
# [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020)
## TL;DR
- 자연어와 이미지를 함께 pre-training한 뒤, downstream task에 zero-shot transfer할 수 있는 방법인 C…
-
Thanks for your work in anomaly detection domain. I am reaching out to discuss an aspect of your work that caught my attention, specifically regarding the experiments conducted in a zero-shot setting.…
-
1. Please check default.yml, if we should add the node of test just like train and val nodes.
![image](https://user-images.githubusercontent.com/25433111/162885280-a2040526-0271-4657-a567-16eafd9381b…
-
[From Visual Prompt Learning to Zero shot transfer ](https://arxiv.org/pdf/2303.05266.pdf)
[Learning to Prompt for Vision-Language Models](https://link.springer.com/article/10.1007/s11263-022-01653…
-
你好,按照我的理解,这个数据的划分是不是测试和训练是不存在重叠的类别,测试是200个训练中没有的类别,那么问题来了,怎么保证能够生成未知的类别的呢?我也看了一下你发到google drive的图像,很明显语义并不是特别的好啊,这个测试训练的划分本质上就不太合理啊,难道不是最理想的状态是我连续10天采集a的脑电信号,头九天是训练,第十天测试吗?
-
Hello,
Could you please let me know where exactly is the "novelty detection" part of the paper "Zero-Shot Learning Through Cross-Modal Transfer" in your code.
-
## 0. Paper
- paper: [link](https://aclanthology.org/2023.acl-long.677/)
- my slide (in Japanese): [link](https://speakerdeck.com/a1da4/wen-xian-shao-jie-whitenedcse-whitening-based-contrastive-…
a1da4 updated
5 months ago
-
Excuse me, I wonder if this is the code of the paper _Zero-shot Sequence Labeling: Transferring Knowledge from Sentences to Tokens_
Thank you so much
-
### Regression?
No
### System Info and Version
System/Version info
```sh
Hyprland, built from branch main at commit 918d8340afd652b011b937d29d5eea0be08467f5 (flake.lock: update).
Date: 2024-…