Chronoberg: 250 Years of English Books Corpus Hits EMNLP Oral

The Chronoberg corpus, spanning 1750–2000 with about 25,000 English books and over 2 billion tokens, was accepted as an EMNLP 2026 oral; it shows LLMs misread historical texts by modern word senses, with data and code to be released as a lifelong-learning testbed.

2026-10-23 ~ 2026-10-23 · 3 related posts

• Chronoberg: EMNLP oral paper releases 250-year English corpus with temporal emotion lexicons to test LLMs — mundt_martin · 2026-10-23
• Chronoberg corpus (EMNLP oral): LLMs misjudge historical text by modern word meanings — mundt_martin · 2026-10-23
• Chronoberg team to present at EMNLP; corpus pitched as a lifelong-learning testbed — mundt_martin · 2026-10-23