{"slug": "timecapsule-generative-hallucination-as-a-method-for-historical-sensemaking", "title": "TimeCapsule: Generative Hallucination as a Method for Historical Sensemaking", "summary": "A 1.2B-parameter LLaMA-style causal model called TimeCapsule, trained exclusively on Victorian texts from 1800-1875, achieves a 45.4% perplexity reduction over a GPT-2 baseline on held-out Victorian prose, according to a new arXiv preprint. The model generates historically plausible analogies for modern concepts, such as describing a computer as a 'hypertrophied lung,' but a qualitative probe with two humanities scholars found that both misclassified approximately 40% of genuine Victorian excerpts as machine-produced. The researchers argue that structural ignorance of the future transforms hallucinations into interpretive probes of nineteenth-century ontologies.", "body_md": "arXiv:2607.24750v1 Announce Type: new\nAbstract: Large Language Models (LLMs) are temporally overexposed: trained on vast contemporary corpora, they encode present-day concepts that make them unreliable narrators of the past. We present TimeCapsule, a 1.2B-parameter LLaMA-style causal model trained exclusively on Victorian texts (1800-1875) as an epistemologically isolated generative archive. Quantitative evaluation shows a 45.4% perplexity reduction over a GPT-2 baseline on held-out Victorian prose, while larger contemporary causal models achieve lower raw perplexity through broader pretraining but lack temporal isolation. TimeCapsule exhibits computational sensemaking, generating historically plausible analogical explanations for unfamiliar modern concepts (e.g., describing a computer as a \"hypertrophied lung\"). A qualitative hermeneutic probe with two humanities scholars revealed a crisis of authenticity, as both misclassified approximately 40% of genuine Victorian excerpts as machine-produced. We argue that structural ignorance of the future transforms hallucinations into interpretive probes of nineteenth-century ontologies.", "url": "https://wpnews.pro/news/timecapsule-generative-hallucination-as-a-method-for-historical-sensemaking", "canonical_source": "https://arxiv.org/abs/2607.24750", "published_at": "2026-07-29 04:00:00+00:00", "updated_at": "2026-07-29 04:26:11.390238+00:00", "lang": "en", "topics": ["large-language-models", "generative-ai", "ai-research"], "entities": ["TimeCapsule", "LLaMA", "GPT-2", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/timecapsule-generative-hallucination-as-a-method-for-historical-sensemaking", "markdown": "https://wpnews.pro/news/timecapsule-generative-hallucination-as-a-method-for-historical-sensemaking.md", "text": "https://wpnews.pro/news/timecapsule-generative-hallucination-as-a-method-for-historical-sensemaking.txt", "jsonld": "https://wpnews.pro/news/timecapsule-generative-hallucination-as-a-method-for-historical-sensemaking.jsonld"}}