Code accompanying the submission "Structural Text Segmentation of Legal Documents" by Aumiller et al.

Extract ESCO skills and ISCO occupations from texts such as job descriptions or CVs

Package to calculate the similarity score between two sentences

COVID-19 Open Research Dataset Challenge (CORD-19)

A python Sentence-Clustering library based on S-Bert and a diverse number of clustering methods. Establishes best number of clusters for each algorithm and the most optimal algorithm by internal and external validation respectively.

Indian Parliamentary Proecedings analysis with NLP

A Python pipeline to generate responses using GPT3, map them to a vector space using the T5 XXL sentence transformer, use PCA and UMAP dimensionality-reduction methods, and then provide visualizations using Plotly and sentiment analysis using TextBlob

Simple Query Based Document Summarization

ML проект для предсказания победителя в CS2 матчах на Faceit.

An off-the-shelf grep-like tool that performs semantic similarity search

Novel multimodal architecture for detecting such misinformation by explicitly modeling the consistency between visual content, textual claims, and external factual knowledge.

Qdrant Vector Database + FastAPI

A semantic meta-search engine and synthesis workstation featuring self-correcting vector memory and local AI insights via Ollama.

A codebaser for Semi-Supervised Word Sense Disambiguation for Ukrainian Sentence Embeddings

download any (almost) book you want

Hybrid AI chatbot for crop diagnostics using RAG, rule-based reasoning, and farm sensor data.

RAG project on fashion dataset. It takes queries from user, classify the queries, and perform retrieval process. In the end, user gets an answer to his question.

Streamlit app that turns any YouTube video into a chat-ready Q&A bot using LangChain, Groq Llama 3.3‑70B, sentence-transformer embeddings, and FAISS for fast semantic search.

A conversational AI chatbot framework built on sentence-transformer model.