Denny Vrandečić<p>Good news: languages that are more widespread have a higher complexity. This means that underserved languages are more likely to be learned well using a smaller corpus, which could help a bit with the rich-get-richer problem of LLMs and existing corpora.</p><p><a href="https://phys.org/news/2025-02-complex-languages-efficient-communication.html#google_vignette" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">phys.org/news/2025-02-complex-</span><span class="invisible">languages-efficient-communication.html#google_vignette</span></a></p><p><a href="https://mas.to/tags/llm" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>llm</span></a> <a href="https://mas.to/tags/smalllanguages" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>smalllanguages</span></a> <a href="https://mas.to/tags/underrepresented" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>underrepresented</span></a> <a href="https://mas.to/tags/underserved" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>underserved</span></a></p>