-
## 一言でいうと
Symmetric Pattern(SP)(たとえばX and Y)に基づく単語ベクトル表現を提案した話。手法としてはSPをコーパスから獲得し、それに基づきPPMIを用いてベクトルを生成する。単語類似度タスクで評価した結果、SimLex999ではSOTAとなった。また、動詞に対して有効であることも分かった。
### 論文リンク
http://www.aclweb.or…
-
Hi, I am interested in the prompt proposed in your paper to get sentence embeddings.
I wonder if I can directly use your prompt in other open-source LLM, e.g. LLAMA 3, to get meaningful sentence embe…
-
I am new to BERT. The main part of BERT is beeing capable of different contextual meaning for a word. But in my case I more need to be able to capture synonyms. So I ask myself if BERT better suited f…
-
Thanks for sharing this work firstly.
I test this code with a reference code, but I got a results as not I expected. As concerned as the similariy it's far away from InstantID performance.
Furtherm…
-
This is not a software issue, we're just wondering whether anyone can shead some light on the results we're seeing.
We've been working on an Icelandic named entity recognizer using NeuroNER. Our tr…
-
**Feature**
Let suppose one has at hand a textual corpus with a split in distinct time periods. One may want to analyze how word embeddings change across time.
**Describe the solution you'd like**…
-
https://blog.reachsumit.com/posts/2020/07/spell-checker-fasttext/
-
Suggest creating a GitHub Actions workflow to run pytest on commits and pull requests.
Steps:
1. Create a new file .github/workflows/test.yml
2. Add configuration to test.yml to configure the e…
-
Word embeddings contain bias. Ethics in general is an active area of research and is worth mentioning.
One of the seminal papers:
* Man is to Computer Programmer as Woman is to Homemaker? Debia…
-
## 一言でいうと
サブワードを分布で表現し、単語をその集合(混合ガウス分布)で表現するという手法(サブワードは文字n-gramで作成する)。式はサブワードの組み合わせのみでなく、単語ベクトルを補完する形で定義されている。単語類似度の精度が上がったほか、未知語対応の効果が確認できた。
![image](https://user-images.githubusercontent.com/…