Pinned Loading

1. pipeline_generation_benchmark pipeline_generation_benchmark Public

   A benchmark consisting of pipelines for sequential coding task execution in VM with configurable data processing components to measure complex reasoning abilities.

   Python

2. bab-ai bab-ai Public

   A WhatsApp based app for pension assistance for seniors in Bulgaria. Created in collaboration with local officials.

   Python

3. seizure-savvy seizure-savvy Public

   An app for helping people with epilepsy understand their seizure patterns and main triggers.

   Python

4. ocr_data_scarcity ocr_data_scarcity Public

   A CNN pipeline for handwriting recognition under conditions of extreme data scarcity (mainly worked with single handedly produced samples)

   Jupyter Notebook 1

5. self-hyde-rag self-hyde-rag Public

   Introducing a novel and improved RAG architecture by replicating and combining approaches outlined in HyDE and SelfRAG papers

   Python

6. LLM_biases LLM_biases Public

   A framework for generating new benchmarking datasets anchoring them in the social science studies. Gender bias evaluation as a proof of concept of the technology

   Jupyter Notebook 1