<?xml version="1.0" encoding="UTF-8"?><urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/sentencepiece-language-agnostic-subword-tokenization</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/why-pretokenization-is-a-problem</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/whitespace-as-a-character-the-prefix</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/training-algorithms</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/a-worked-example</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/code-implementation</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/subword-regularization-in-depth</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/multilingual-vocabulary-allocation</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/limitations-and-impact</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/summary</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url><url><loc>https://mbrenndoerfer.com/writing/sentencepiece-subword-tokenization-bpe-unigram/quiz</loc><lastmod>2025-04-14T00:00:00.000Z</lastmod></url></urlset>