Deployments of Polish LLMs in Administration and the Private Sector
summary
Presentation covers development and deployment of Polish large language models (PLLuM) by the HIVE AI consortium, detailing data preparation, training methods, evaluation, and methodological choices. It presents new PLLuM model variants, approaches for securing and aligning them to user needs, introduces an improved ShpaRAG tool for designing and testing single- and multi-turn RAG systems, and shares experience deploying Polish LLMs in public administration and private sector.
AI-written summary · about
chapters (4)
- 0:00HFI project and Plum models
- 10:55Language data and Plum models
- 18:39Plum deployments in public and private sector
- 24:56Q&A
related talks
49mA Practical Approach to Using LLMs: Peculiarities and Pitfalls
33mDesigning LLM Applications for Massive Datasets
43mIntegrating AI with RPA
30mWhat Can LLMs Do with Messy Data?
47mA Step-by-Step Guide to Securing Large Language Models (LLMs)
25mLLM Powered Agents with Kubernetes
15mLLMs for Data-Driven Decision Making in Product Development
41mBuilding LLM Agents – Challenges and Learnings