Beyond English: Challenges of Building a Ukrainian-Language RAG System
summary
Examines challenges of building a Retrieval-Augmented Generation (RAG) system for Ukrainian, focusing on evaluation methods in domain-specific settings such as public services consulting. Covers practical implementation issues and common pitfalls encountered when deploying Ukrainian-language RAG systems.
AI-written summary · about
chapters (15)
- 0:00RAG systems intro
- 2:37Four RAG architecture reasons
- 5:05RAG use cases
- 6:09Five RAG pipeline stages
- 9:22Indexing: data prep
- 11:56Retrieval: relevant data search
- 14:05Augmentation & generation
- 15:31RAG quality issues
- 16:14RAG evaluation metrics
- 22:57Ukrainian in RAG
- 24:01Ukrainian benchmark challenges
- 26:36Embeddings: expectations vs reality
- 28:01Role of Ukrainian data
- 30:52Future of Ukrainian RAG
- 31:57Summary & open questions
related talks
51mDeveloping a Custom RAG System
38mRetrieval-Augmented Generation (RAG) in NLP
44mWhat Is a RAG System and How to Build One
19mImproving RAG Systems - Key Challenges and Solutions
53mFrom RAG to Riches: Secrets of Data-Augmented LLMs
01h 28mBuilding a RAG System on Node.js
25mAdvanced RAG Systems: Scaling Retrieval-Augmented Generation in Production
01mtalk about RAG