Viewing profile — rezamoaiandin

A joint study by Anthropic and the UK AI Security Institute found that bad actors only need to contaminate a dataset with around 250 malicious documents to introduce a specific bac…

Yes! it's a great fit for this problem, we actually experimented with it: https://medium.com/sculpt/a-technique-for-building-nlp-class...

SculptAI is currently available to selected customers of SALT. But happy to discuss if you would like access. Please get in touch with us

We used a shallow neural network. The main challenge at Sculpt wasn't the modelling part, but rather the whole UX (active learning to speed up the training process, explainability,…

Given that is of interest, I'll do a follow up on the implementation details and dataset!

Thats pretty amazing!