New answers tagged

2 votes

Public Email Classification Dataset but not Spam vs Ham

If I am understanding correctly, you want to create a model that takes an email body and assigns some probability to a pre-specified set of classes (feedback, complaint, lost and found, etc.) ...
AndrewJaeyoung's user avatar
4 votes

How do I prompt GPT-4 to look at a PDF in Jupyter Notebook?

OpenAI's API does not support providing PDFs. Therefore, what you intend to do is not possible. You have other options though: Parse the PDF locally and send text to GPT4. The parse may consist of ...
noe's user avatar
  • 26.6k
0 votes

State-of-the-art Python packages that can evaluate language similarity

GPT-based evalualtion are often mentioned as SOTA currently. E.g., one can use (paper: "G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment" [...
Franck Dernoncourt's user avatar
1 vote

Using doccano for Aspect Based Sentiment Analysis annotation

Aspect Based Sentiment Analysis often involves linking aspect terms with polarity terms, e.g. from Since 2022, Doccano also supports relation annotation: To ...
Franck Dernoncourt's user avatar
0 votes

Recommended way to embed a text thousands of tokens long?

You may use different ways to combine them together, but averaging would be the simplest where you may loose some details and enhance common concepts in the list that you are combining. I would ...
Payam Jome Yazdian's user avatar
0 votes

English to "basic English" translation

Text simplification is not as popular as other generative tasks (e.g. summarization, translation). Nevertheless, you may try the approach of the Keep It Simple: Unsupervised Simplification of Multi-...
noe's user avatar
  • 26.6k
0 votes

Transformer model: Why are word embeddings scaled before adding positional encodings?

Multiplying Weights by √dmodel In the embedding layers, the weights are multiplied by the square root of dmodel. This is done to scale the weights appropriately and ensure that the dot product between ...
dreamg's user avatar
  • 1
3 votes

Why does everyone use BERT in research instead of LLAMA or GPT or PaLM, etc?

Adding/complementing the other answers, BERT gives the possibility to access/obtain the embeddings of the fed input (which wasn't and still isn't the case of some other models). The embeddings are ...
ahmed_khan_89's user avatar
0 votes

Question about contextual embeddings?

Start with a sequence of embeddings. In the standard attention computation, each embedding in the sequence attends to every other embedding in the sequence. This can be considered "bidirectional&...
Karl's user avatar
  • 636

Top 50 recent answers are included