Alexander Wan's user avatar
Alexander Wan's user avatar
Alexander Wan's user avatar
Alexander Wan
  • Member for 7 months
  • Last seen this week
8 votes

Do neural network weights need to add up to one?

3 votes

Does Negative Prompting Exist?

3 votes
Accepted

Aren't context lengths for transformers an artificial restriction?

3 votes
Accepted

How to train a sample weight model for another ML model?

3 votes

Does Negative Prompting Exist?

2 votes

Empty space detection

2 votes

Match two paragraphs of text

2 votes

Noob crafting a simple "Zero-Shot Classifier" Using an API . How can I avoid passing the categories every single request?

2 votes

Are popular songs considered outliers to AI

2 votes
Accepted

Is it possible to reconstruct convolutional layers' input using transposed convolution?

2 votes
Accepted

Fine Tuning a Bert Transformer. How to label for emotions and train large scripts?

2 votes

What is the best Text-to-speech model available open-source?

2 votes
Accepted

What is $\mathbf{S}$ (sample covariance matrix) in image compression based on PCA?

2 votes
Accepted

How the "quantizantion" of models does work?

2 votes

Why can't Lucene search be used to power LLM applications?

2 votes
Accepted

Why does the latent space in Stable Diffusion have a shape of 64x64x3?

1 vote

Why use learning rate schedules if weight updates automatically decrease when approaching local optimal?

1 vote

Handling Variable Output Token Dimensionality in Transformer Decoders During Inference

1 vote
Accepted

Why is generative QA + RAG so popular when extractive QA models exist that can provide factual information?

1 vote
Accepted

Attention with Recurrent Neural Networks

1 vote
Accepted

Why does skip-gram uses linear maps for embedding?

1 vote
Accepted

In the paper "LLM in a flash," what is meant by an up projection or down projection layer?

1 vote

How to Create a 1D Embedding from Tensors of Varying Sizes?

1 vote
Accepted

Overcoming the quadratic scaling in transformer architecture

1 vote
Accepted

Navigating encoder-only text/sentence embedding models

1 vote

What does "fully visible belief network" stand for?

1 vote

Can we generate labels for an unlabelled dataset by doing some feature engineering?

1 vote

Do I need to manually segment/augment my data for transformer training?

1 vote

Number of units in Final softmax layer in VGGNet16

1 vote

How can BERT/Transformer models accept input batches of different sizes?