Faculty of Electrical Engineering and Computing in Zagreb

Assessing Graph Data Serialization and Context Selection for Knowledge-Augmented Question Answering with Large Language Models

Large language models (LLMs) can produce inaccurate answers even when relevant facts are supplied in the prompt. This study examines how graph data serialization, entity identifiers, and context selection affect factual question answering in knowledge-augmented prompting. Four sy...

FEDCODE: A Framework and Experimental Study of Model-Update Compression for Communication-Efficient Federated Learning

Federated learning trains shared models without centralizing local data, but repeated model exchange can become a bottleneck in bandwidth-, energy-, and latency-constrained edge systems. This paper presents FEDCODE, a communication-aware federated learning simulation framework fo...