Victorian-era LLM 'TimeCapsule' generates historically plausible text

Researchers have developed TimeCapsule, a 1.2 billion parameter LLaMA-style model trained exclusively on Victorian-era texts from 1800-1875. This model aims to provide a historically isolated generative archive, demonstrating a 45.4% perplexity reduction compared to a GPT-2 baseline on Victorian prose. While larger contemporary models achieve lower perplexity, they lack temporal isolation. TimeCapsule can generate historically plausible explanations for modern concepts, and humanities scholars found it difficult to distinguish its output from genuine Victorian writing. AI

IMPACT This research explores methods for creating temporally isolated LLMs, potentially improving historical analysis and understanding of past ontologies.

RANK_REASON The cluster describes a new research paper detailing a novel LLM trained on a specific historical corpus. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• GPT-2
• Hugging Face
• LLaMA
• Victorian

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →