-
||link|
|----|---|
|paper| [Cross Modal Retrieval with Querybank Normalisation](https://arxiv.org/pdf/2112.12777v3.pdf) |
|code| [papers with code](https://paperswithcode.com/paper/cross-modal-retr…
-
When I used ut_medium to do experiments on the finetune kinetics400 dataset, I found that categorical_cross_entropy decreased at the beginning, began to rise after 3000steps, and top1 began to decline…
-
One issue that @eric-czech has noticed during error analysis for relation extraction: it's hard to labeling functions that exclude negated relations.
It may be worth looking at how others have appr…
-
-
Please post here interesting papers, possibly with a brief description of technique and results (improvement).
-
# Summary
기존의 VLP는 from scratch로 학습을 시켰지만, 이는 pre-training cost가 너무 크며 기존에 잘 학습되었던 모델 (특히, LLM)에 대한 활용이 어려움. 따라서, frozen vision encoder와 frozen llm을 Q-Former (Querying Transformer)를 통해 잘 이어보는 방식으…
-
https://arxiv.org/pdf/1702.05464.pdf
Adversarial learning methods are a promising approach to training robust deep networks, and can generate complex samples across diverse domains. They also can i…
leo-p updated
7 years ago
-
-
Hi, thanks for the great work!
I note that in the linear evaluation setting, the fusion results of the proposed method achieve huge gains in NTU 120 (7+% gains), which is not observed in NTU 60. Mea…
-
## Why
Machine Learning 輪講は最新の技術や論文を追うことで、エンジニアが「技術で解決できること」のレベルをあげていくことを目的にした会です。
prev. #34
## What
話したいことがある人はここにコメントしましょう!
面白いものを見つけた時点でとりあえず話すという宣言だけでもしましょう!