Building a Swedish Open-Domain Conversational Language Model

TL;DR

This paper introduces a large Swedish conversational language model trained on online forum data, demonstrating promising human-like and informative responses, while also discussing ethical considerations and safety measures.

Contribution

The paper presents the first large Swedish conversational model trained on online forum data and evaluates its conversational abilities through human assessment.

Findings

Model responds in a human-like manner

Model covers diverse topics effectively

Highlights ethical considerations in deployment

Abstract

We present on-going work of evaluating the, to our knowledge, first large generative language model trained to converse in Swedish, using data from the online discussion forum Flashback. We conduct a human evaluation pilot study that indicates the model is often able to respond to conversations in both a human-like and informative manner, on a diverse set of topics. While data from online forums can be useful to build conversational systems, we reflect on the negative consequences that incautious application might have, and the need for taking active measures to safeguard against them.