Humanities · OECD FoS 6.1 · China MoE 0601 考古学
Archaeology & classics
考古学与古典文献
L2Local breakthroughs ↑ Rising Score 1.90 · Last reviewed 2026-10-22 · Next review 2027-01-19
BIGGEST BREAKTHROUGH 2026-06-25 Field leapAI · major
Vesuvius Challenge: first entire Herculaneum scroll read (PHerc. 1667, ~20 columns, ~1.5 m) ↗
A carbonised 2,000-year-old scroll read in full without opening
Why this level
Ithaca (Nature 2022) and Aeneas (Nature 2025) restore and date inscriptions; the Vesuvius Challenge read carbonised Herculaneum papyri (2024 prize; ~1.5 m of Greek from PHerc. 1667 in 2026); AI helped find 303 new Nazca geoglyphs (PNAS 2024).
S = 0.30·method + 0.35·discovery + 0.25·adoption + 0.10·education (education scores 0 when there is no evidence; if empty, the other weights are renormalised)
Sources
- nature.com/articles/s41586-022-04448-z
- nature.com/articles/s41586-025-09292-5
- scrollprize.org/grandprize
- pnas.org/doi/10.1073/pnas.2407652121
Composite 1.5–2.49: at least one peer-reviewed/officially verified AI-key result, or one dimension at 3; routine workflow largely unchanged. All levels →
Research directions
Inscription restoration & datingVirtual unwrapping & ink detectionRemote-sensing site discoveryArtefact refitting
Papers
- Beyond site-specific archaeological prospection: Spectral-geometry reveals cross-site cropmark signals in satellite data
AI: constrained SVM/XGBoost with a directional strength score explains transfer. Field: challenges the assumption that cropmark detection is site-specific, supporting regional satellite prospection (three sites)
2026-10-10 · PNAS Nexus
- Contextualizing ancient texts with generative neural networks
Latin inscription contextualisation
2025-07-23 · Nature
- AI-accelerated Nazca survey nearly doubles the number of known figurative geoglyphs and sheds light on their purpose
303 new geoglyphs, nearly doubling the known count
2024-09-23 · Proceedings of the National Academy of Sciences
- Restoring and attributing ancient texts using deep neural networks
62% restoration accuracy, 72% with historians
2022-03-09 · Nature
- Design and cultural heritage in architecture in the age of artificial intelligence: generative models, HBIM, and evidence-based digital reconstruction
Argues generative-AI outputs become heritage knowledge only when validated against historical evidence and conservation constraints in HBIM.
2026-10-19 · OpenAlex · Virtual Archaeology Review · Ömer Faruk Pamak, Cemile Feyzan Şimşek
- An evidence-informed fine-tuning framework for traceable generative redesign of architectural heritage ornaments
Evidence-informed fine-tuning converts audited heritage evidence into conditions for generative ornament redesign.
2026-10-17 · OpenAlex · npj Heritage Science · Jiale Zhang, Hanbin Luo, Jie Xu
- Reconstructing Lost Urban Memory with Generative AI: An Oral-History Prompt-Engineering Framework
Turns oral histories into prompts for generative AI to visualise a lost Kayseri neighbourhood.
2026-10-16 · OpenAlex · Turkish Journal of Engineering · Özlem Kevseroğlu, İrem Yaşar, Vildan Yıldırım
- Digital preservation and automated roof damage detection for traditional village heritage in Guangdong using UAV photogrammetry and deep learning
UAV photogrammetry plus YOLOv8 (mAP50 0.832) detects roof damage in five Guangdong heritage villages.
2026-10-16 · OpenAlex · npj Heritage Science · ZHU Zhengyang, Chun-Ming Hsieh, Xiaodong Liu, LI Xiangsi
- From fragmented narratives to spatio-narrative networks: a knowledge graph approach to Tianjin Grand Canal heritage
LLM-assisted extraction builds a narrative knowledge graph of 98 Grand Canal heritage sites in Tianjin.
2026-10-16 · OpenAlex · npj Heritage Science · Bolun Zhang, Jingwen Tian, Lihao Meng, Jiaying Li
- InscriptionOCR: A Dataset and Method for Understanding Inscriptions
End-to-end AI pipeline for Ashokan Brahmi stone inscriptions: restoration, OCR, transliteration and translation.
2026-10-07 · arXiv · Jaidev Sanjay Khalane, Akbar Ali, V. N. Prabhakar, Shanmuganathan Raman
- From arXiv, bioRxiv, medRxiv and OpenAlex (last ~30 days); summaries written by AIMap from the abstract.
Milestones 7· 2 leaps
Big glowing nodes are leaps: AI leap cool · Field leap warm · double ring = both
-
Paper
PNAS Nexus: Sentinel-2 cropmark detection models trained at one archaeological site transfer to others, governed by spectral-geometric similarityAI: constrained SVM/XGBoost with a directional strength score explains transfer. Field: challenges the assumption that cropmark detection is site-specific, supporting regional satellite prospection (three sites)
PNAS Nexus -
Discovery
AI and X-rays recover ~1.5 m of Greek text from PHerc. 1667Media coverage of the same result
CNN -
DiscoveryField leap
Vesuvius Challenge: first entire Herculaneum scroll read (PHerc. 1667, ~20 columns, ~1.5 m)A carbonised 2,000-year-old scroll read in full without opening
Vesuvius Challenge -
Paper
Aeneas: contextualising and restoring Latin inscriptions (Nature)Latin inscription contextualisation
Nature -
Paper
AI helps find 303 new Nazca geoglyphs (PNAS)303 new geoglyphs, nearly doubling the known count
PNAS -
CompetitionField leap
Vesuvius Challenge Grand Prize: AI reads Herculaneum papyrus passagesFirst passages read from an unopened carbonised scroll
Vesuvius Challenge -
Paper
Ithaca restores and dates Greek inscriptions (Nature)62% restoration accuracy, 72% with historians
Nature
No milestones match this filter.
More papers & repos 5
| Type | Title | Date |
|---|---|---|
| Paper | PyPottery (arXiv 2610.02072) PreprintTo be verified | 2026-10 |
| Paper | Contextualizing ancient texts with generative neural networks | 2025-07-23 |
| Paper | AI-accelerated Nazca survey nearly doubles the number of known figurative geoglyphs and sheds light on their purpose | 2024-09-23 |
| Paper | Restoring and attributing ancient texts using deep neural networks | 2022-03-09 |
| Repo | google-deepmind/ithaca To be verified | — |
Latest news 5
- Scientists decipher charred ancient scrolls buried by VesuviusThe Irish Timesen
- NYU Researchers Are Building AI to Piece Together Broken Archaeological FindsNYUenTo be verified
- Bringing ancient Mesopotamia into the AI ageGeorge Mason UniversityenTo be verified
- First entire Herculaneum scroll readVesuvius Challengeen