AI in Production

Generative (TrOCR) models calibration

Teaching a generative TrOCR model to know what it doesn't know

Dec 13, 2023 • 

Tin Ferković

3

November 2023

LinguaAI: AI-Powered Personalized Conversations for Comprehensive Language Learning

Beyond Traditional Learning: The GPT-4 Edge in Adaptive Language Tutoring for any Language and any Level.

Nov 2, 2023 • 

Mihaela Bakšić

4

August 2023

Multi-Task Learning with Intermediate Continual Learning for Industry NLP Use Cases

Utilizing adapters and hypernetworks to efficiently, effectively, and continuously train multiple tasks

Aug 21, 2023 • 

Tin Ferković

2

Exploring State-of-the-Art Models for Document-Level Entity Relation Extraction

Document-level entity relation extraction involves identifying and categorizing relationships between entities mentioned within whole documents, rather…

Aug 8, 2023 • 

Dunja Smigovec

2

July 2023

Named Entity Recognition Using Question Answering in Zero- and Few-Shot Settings

The human race produces copious amounts of documents daily to aid its effective functioning.

Jul 28, 2023 • 

Luka Pavlović

1

Secure Your Kubernetes Web Application Using Gatekeeper And Keycloak

In certain scenarios, you may need to integrate an authentication layer into a basic application or tool that doesn't demand complex authentication…

Jul 3, 2023 • 

Bojan

5

May 2023

Data Extraction from Invoice Type Documents using Large Language Models

Towards tailored made AI models that adapt to changes and match the needs and demands of a dynamic real-world environment.

May 23, 2023 • 

Mihaela Kakarigi

 and 

zeljko

15

January 2023

Annotation Team Lead and Teamwork

Document annotation is a crucial step in training AI models, particularly in the context of Natural Language Processing (NLP).

Jan 13, 2023 • 

Dragan Kraljević

11

1

Named Entity Recognition Using a Pre-Trained LayoutLMv2 Model

The amount of data generated by humans is ever-increasing, much of which is stored in documents and forms that need to be analyzed to extract relevant…

Jan 11, 2023 • 

Marko Cvjetko

11

Named Entity Recognition using BERT on Subsampled Datasets

The Named-Entity Recognition (NER) system I worked on during my internship at doXray is part of a live production system.

Jan 11, 2023 • 

Eva Jagodić

 and 

Bruno Šarlija

11

1