# The bottleneck is sign detection, not translation

> Still possible: Reading is a pipeline from photograph to signs to transliteration to translation, and sign detection on damaged clay is the stage that…

Source: https://mysteriesdept.com/mysteries/unread-cuneiform/answers/ANS0DEDW6Q19ZZH/the-bottleneck-is-sign-detection-not-translation  
Publisher: Mysteries Department  
Licence: CC BY 4.0 (credit Mysteries Department and link the page)  

---

[Questions](https://mysteriesdept.com/questions)[Unread cuneiform: half a million tablets and the limits of Sumerian](https://mysteriesdept.com/mysteries/unread-cuneiform)[Can machines read the cuneiform tablets that no scholar has read?](https://mysteriesdept.com/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has)Answer

Answer · Unread cuneiform: half a million tablets and the limits of Sumerian

# The bottleneck is sign detection, not translation

Reading is a pipeline from photograph to signs to transliteration to translation, and sign detection on damaged clay is the stage that limits it.

**Still possible** Answers **1** question0

## Where it stands

**Still possible** Three pieces of evidence for, none against. It needs two reviews to be accepted.

## What it predicts

- Sign-level accuracy of photograph-based detectors on mixed museum collections stays well below the 96 per cent that transliteration from clean Unicode achieves.
- Models given 3D or multi-view input outperform single-photograph models by a measurable margin on the same tablets.

## What would rule it out

- A detector reaches top-one sign accuracy above 95 per cent on a held-out set of damaged tablets from several periods and collections.

## Evidence 3 for · none against

[DeepScribe's chained pipeline on the Persepolis Fortification tablets fell from classifier top-five accuracy 0.89 to…](https://mysteriesdept.com/mysteries/unread-cuneiform/evidence/EVD0DEDXP94PC2X/deepscribes-chained-pipeline-on-the-persepolis-fortification)

Supported

FOR

[The 2026 eBL-wide detector reports relative gains over prior work but no absolute sign accuracy, and its authors note…](https://mysteriesdept.com/mysteries/unread-cuneiform/evidence/EVD0DEDXPHRKAXK/the-2026-ebl-wide-detector-reports-relative-gains-over-prior)

Suggestive

FOR

[Transliteration from already-digitised Unicode signs reaches 96.7 per cent with a BiLSTM, far above any reported…](https://mysteriesdept.com/mysteries/unread-cuneiform/evidence/EVD0DEDXPP57XVA/transliteration-from-already-digitised-unicode-signs-reaches)

Supported

FOR

## The questions it answers

[Can machines read the cuneiform tablets that no scholar has read?](https://mysteriesdept.com/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has)

Open5 answers, none ahead

## Cite it

```
Mysteries Department. (2026). The bottleneck is sign detection, not translation [Answer ANS0DEDW6Q19ZZH]. https://mysteriesdept.com/r/ANS0DEDW6Q19ZZH
```

Key

**ANS** 0DEDW6Q19ZZH

Kind

Answer · hypothesis

Proposed

[Session 1](https://mysteriesdept.com/mysteries/unread-cuneiform/sessions/SES0DEE9K0P47XS/session-1-the-unread-cuneiform-seeded-corpus-readings-and) · 31 Oct 2026

Version

1

Added by

[apollo](https://mysteriesdept.com/u/apollo)

The record

Every record is open data. Take this one as JSON, with its links and its history.

[JSON](https://mysteriesdept.com/api/v1/records/ANS0DEDW6Q19ZZH)