gpt-5.5_v4_reasoning v1 Evaluation

Generated: 2026-09-20 11:22:28 UTC

Pairs100
Distinct lemmas100
Slope0.916
R^20.959
BLEU-455.6%
chrF++75.0%
METEOR80.4%
ROUGE-L80.6%
BERTScore96.6%
COMET80.4%
BLEURT77.8%

Profile version id: 1745

First translation using this version: 2026-06-21

Total usable AI runs for this version: 101 runs across 101 lemmas

Created: 2026-06-19 00:36:32.588791+10:00

Active: True

Notes: Approved-human-only v4 reasoning trial using Responses API reasoning.effort=low.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.916
Intercept1.818
R^20.959
P-value1.25e-69
Pearson r0.979
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words42.270
Mean absolute length residual3.515
Mean absolute percent error7.9%

Metric vs Passage Length

These plots test whether score changes with source passage length for this prompt version. The dashed line is an ordinary least-squares fit; the annotation reports n, r, R^2, p-value, and slope.

BLEU-4 vs passage length
BLEU-4: significant negative; r = -0.403, R^2 = 0.163, p = 3.19e-05.
chrF++ vs passage length
chrF++: significant negative; r = -0.408, R^2 = 0.166, p = 2.57e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.397, R^2 = 0.157, p = 4.39e-05.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.387, R^2 = 0.150, p = 7.08e-05.
BERTScore vs passage length
BERTScore: significant negative; r = -0.478, R^2 = 0.228, p = 4.98e-07.
COMET vs passage length
COMET: significant negative; r = -0.397, R^2 = 0.157, p = 4.40e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.676, R^2 = 0.457, p = 1.18e-14.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.334, R^2 = 0.111, p = 0.0007.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.339, R^2 = 0.115, p = 0.0006.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.337, R^2 = 0.114, p = 0.0006.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.342, R^2 = 0.117, p = 0.0005.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.403 0.163 3.19e-05 -0.00295 ok
chrF++ 100 significant negative -0.408 0.166 2.57e-05 -0.00206 ok
METEOR 100 significant negative -0.397 0.157 4.39e-05 -0.00193 ok
ROUGE-L 100 significant negative -0.387 0.150 7.08e-05 -0.00184 ok
BERTScore 100 significant negative -0.478 0.228 4.98e-07 -0.00047 ok
COMET 100 significant negative -0.397 0.157 4.40e-05 -0.00102 ok
BLEURT 100 significant negative -0.676 0.457 1.18e-14 -0.00275 ok
Trigram precision 100 significant negative -0.334 0.111 0.0007 -0.00258 ok
Trigram recall 100 significant negative -0.339 0.115 0.0006 -0.00254 ok
Trigram F1 100 significant negative -0.337 0.114 0.0006 -0.00255 ok
Trigram Jaccard 100 significant negative -0.342 0.117 0.0005 -0.00277 ok

Translation Similarity Metrics

Mean BLEU-455.6%
Mean chrF++75.0%
Mean METEOR80.4%
Mean ROUGE-L80.6%
Mean BERTScore96.6%
Mean COMET80.4%
Mean BLEURT77.8%
Legacy smoothed corpus BLEU53.6%
Unigram F180.7%
Bigram F160.5%
Trigram precision48.8%
Trigram recall47.3%
Trigram F148.1%
Trigram Jaccard31.6%
4-gram F138.9%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
7267 Κωνώπη 64 2 -58.458 0.0% 0.7% 0.0% 0.0% 78.7% 33.5% 26.3% reviewed/approved 20358 gpt-5.5
2604 Καρχηδών 117 133 23.981 33.9% 65.9% 75.3% 71.4% 93.8% 78.4% 62.6% reviewed/approved 20228 gpt-5.5
2484 Καρία 270 231 -18.205 44.2% 66.2% 65.8% 71.1% 92.8% 80.5% 53.7% reviewed/approved 19732 gpt-5.5
2603 Κάρυστος 175 176 13.839 49.9% 70.8% 77.6% 76.9% 95.5% 78.9% 63.9% reviewed/approved 20227 gpt-5.5
3288 Κορώνεια 110 116 13.395 41.9% 71.2% 69.3% 74.0% 95.1% 77.0% 69.1% reviewed/approved 20265 gpt-5.5
2342 Κάμιρος 54 61 9.705 53.0% 78.7% 84.3% 78.3% 96.3% 80.9% 73.7% reviewed/approved 18403 gpt-5.5
2465 Κάνωπος 83 86 8.134 45.7% 70.7% 81.3% 78.1% 96.1% 78.2% 75.5% reviewed/approved 19728 gpt-5.5
7255 Κυτέριον 22 30 8.025 51.1% 76.9% 80.4% 73.1% 97.0% 78.2% 80.0% reviewed/approved 20324 gpt-5.5
7260 Κυχρεῖος πάγος 61 65 7.291 48.6% 71.0% 78.7% 68.3% 96.4% 77.4% 67.0% reviewed/approved 20351 gpt-5.5
2468 Καπετώλιον 138 121 -7.260 21.0% 48.0% 53.7% 61.0% 88.8% 69.4% 54.3% reviewed/approved 19729 gpt-5.5
7259 Κύφος 54 58 6.705 43.8% 73.1% 90.7% 85.7% 96.0% 76.5% 80.1% reviewed/approved 20328 gpt-5.5
2056 Καβειρία 112 98 -6.438 42.9% 72.0% 69.5% 73.3% 96.3% 79.9% 61.0% reviewed/approved 8958 gpt-5.5
2605 Κάσιον 56 59 5.872 56.9% 77.2% 86.7% 80.0% 96.7% 78.6% 74.0% reviewed/approved 20229 gpt-5.5
3496 Κοτιάειον 71 61 -5.871 27.6% 53.7% 55.1% 60.6% 93.1% 73.6% 64.5% reviewed/approved 20266 gpt-5.5
2328 Καλλίπολις 43 47 5.784 45.4% 67.3% 75.6% 71.1% 95.1% 75.2% 72.1% reviewed/approved 18395 gpt-5.5
2470 Καππαδοκία 68 69 4.877 17.2% 55.0% 81.5% 77.4% 95.3% 76.8% 72.5% reviewed/approved 19730 gpt-5.5
2477 Καρδαμύλη 43 46 4.784 36.2% 67.7% 84.9% 74.2% 96.5% 80.3% 73.7% reviewed/approved 19731 gpt-5.5
7264 Κωλιάς 61 62 4.291 56.6% 76.3% 83.4% 79.7% 96.8% 75.5% 70.2% reviewed/approved 20355 gpt-5.5
2062 Κάθαια 25 29 4.276 23.0% 61.0% 69.4% 55.6% 95.6% 79.8% 77.5% reviewed/approved 8965 gpt-5.5
2621 Κάστνιον 25 29 4.276 55.2% 78.1% 77.7% 77.8% 97.5% 82.1% 77.1% reviewed/approved 20233 gpt-5.5
7254 Κύτα 84 83 4.217 49.2% 67.2% 79.5% 75.4% 94.5% 77.5% 65.4% reviewed/approved 20323 gpt-5.5
2083 Καλαμένθη 24 20 -3.808 44.4% 64.3% 72.9% 77.3% 94.7% 77.4% 75.6% reviewed/approved 13741 gpt-5.5
2455 Κάναστρον 60 53 -3.793 25.4% 57.5% 58.9% 65.5% 93.7% 71.7% 62.8% reviewed/approved 19727 gpt-5.5
2331 Κάλυδνα 43 45 3.784 67.0% 86.1% 94.6% 93.2% 98.1% 82.7% 83.4% reviewed/approved 18398 gpt-5.5
7258 Κύτωρος 30 33 3.695 60.8% 84.9% 90.3% 85.7% 97.6% 82.2% 81.9% reviewed/approved 20327 gpt-5.5
7253 Κυρτώνιος 18 22 3.690 43.4% 66.7% 73.1% 70.0% 96.6% 77.9% 81.8% reviewed/approved 20322 gpt-5.5
2597 Καρπασία 112 108 3.562 44.1% 68.4% 75.5% 80.0% 95.3% 76.8% 64.0% reviewed/approved 19733 gpt-5.5
2114 Κάλβιος 18 15 -3.310 56.9% 72.8% 72.4% 78.8% 95.1% 76.0% 77.4% reviewed/approved 16556 gpt-5.5
7261 Κύψελα 34 30 -2.970 61.8% 77.3% 78.9% 81.2% 97.9% 81.6% 77.6% reviewed/approved 20352 gpt-5.5
7248 Κύρου πόλις 32 34 2.862 40.2% 68.7% 73.2% 72.7% 96.1% 81.3% 79.9% reviewed/approved 20296 gpt-5.5

Residual Predictors

Positive signed-residual terms are associated with AI translations longer than expected from the human length; negative signed-residual terms are associated with shorter AI translations. Absolute-residual models identify terms associated with larger AI-human length divergence.

Greek source terms predicting signed length residuals

Features: 176; ridge alpha: 17.013; cross-validated R^2: -0.087.

Positive termCoefficientDocsMean presentMean absent
απο 0.7240 27 3.123 -1.155
δε και 0.6867 8 5.517 -0.480
εστι δε και 0.5494 3 14.053 -0.435
πολιτης 0.5474 18 4.275 -0.938
ου 0.4802 14 3.114 -0.507
εστι δε 0.4607 5 7.253 -0.382
εκαλειτο 0.4338 7 3.301 -0.248
παιδος 0.3905 4 5.972 -0.249
πολις 0.3709 68 0.287 -0.611
ος 0.3709 6 4.539 -0.290
εκαλειτο δε 0.3646 3 6.538 -0.202
οι 0.3616 11 3.848 -0.476
και το 0.3458 9 4.064 -0.402
δυναται 0.3065 3 7.910 -0.245
ως 0.2828 52 0.820 -0.889
εν 0.2821 37 1.674 -0.983
δε 0.2787 46 0.292 -0.249
πολις και 0.2781 6 5.525 -0.353
ου και 0.2740 4 7.051 -0.294
τη 0.2725 15 1.290 -0.228
Negative termCoefficientDocsMean presentMean absent
τους -1.5649 4 -14.028 0.584
πολυβιος -0.7484 5 -11.387 0.599
φασι -0.7282 8 -5.340 0.464
τα -0.6149 12 -5.065 0.691
το -0.6053 70 -0.400 0.933
παρα -0.6015 7 -8.894 0.669
το εθνικον -0.5934 56 -0.926 1.178
εθνικον -0.5934 56 -0.926 1.178
εστι και -0.4727 20 -1.548 0.387
και -0.2832 66 0.153 -0.297
οικητωρ -0.2805 6 -4.774 0.305
γαρ -0.2266 9 -0.220 0.022
θρακης -0.2116 5 -2.287 0.120
των -0.2071 17 -1.192 0.244
εκαταιος ασια -0.1990 5 -1.557 0.082
ασια -0.1990 5 -1.557 0.082
εθνος -0.1893 8 -3.196 0.278
τινες δε -0.1793 3 -3.607 0.112
την -0.1736 13 -2.009 0.300
πολις θρακης -0.1720 4 -1.911 0.080

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 5.878; cross-validated R^2: -0.782.

Positive termCoefficientDocsMean presentMean absent
τους 3.9055 4 16.516 2.974
και 3.7848 66 4.532 1.542
την 2.8098 13 12.508 2.172
παρα 2.1520 7 14.849 2.662
φασι 1.7348 8 12.637 2.722
πολυβιος 1.5932 5 12.906 3.021
δε 1.5432 46 5.522 1.806
τα 1.5169 12 9.994 2.632
εστι 1.4275 24 7.385 2.293
δε και 1.2343 8 10.707 2.890
εκαλειτο δε 1.1057 3 18.675 3.046
εστι δε και 0.9816 3 14.053 3.189
των 0.9617 17 8.104 2.575
της 0.9448 35 6.035 2.159
εστι και 0.8914 20 7.075 2.625
ος 0.8557 6 8.916 3.171
εκαλειτο 0.8267 7 8.922 3.108
εστι δε 0.7162 5 9.828 3.183
εκαλειτο δε και 0.7134 3 18.675 3.046
δια 0.6481 16 6.418 2.962
Negative termCoefficientDocsMean presentMean absent
ως -1.3349 52 3.119 3.945
στραβων -0.5902 15 3.416 3.533
νησος -0.5494 9 2.409 3.625
προς τη -0.4787 4 1.437 3.602
οτι -0.4585 5 2.017 3.594
προς -0.4565 15 3.484 3.521
πορρω -0.4491 3 0.140 3.620
ου πορρω -0.4491 3 0.140 3.620
εκ -0.4436 6 2.100 3.606
μια -0.4282 3 1.542 3.576
πλησιον -0.4281 8 2.360 3.616
τεταρτω -0.4222 3 0.834 3.598
τον -0.3966 10 3.365 3.532
οι -0.3929 11 4.098 3.443
και του -0.3873 4 1.096 3.616
του δε -0.3791 3 1.230 3.586
και πολις -0.3699 5 3.044 3.540
τινες -0.3646 6 2.535 3.578
ης -0.3513 3 2.781 3.538
κωμη -0.3456 3 2.509 3.546

AI English terms predicting large absolute residuals

Features: 332; ridge alpha: 24.245; cross-validated R^2: -1.045.

Positive termCoefficientDocsMean presentMean absent
with 0.2134 16 5.525 3.132
there is 0.2114 24 5.485 2.893
there 0.2102 32 4.977 2.828
he 0.1899 4 13.837 3.085
be 0.1858 13 7.139 2.974
also 0.1828 42 4.583 2.742
is named 0.1752 4 14.627 3.052
is named after 0.1752 4 14.627 3.052
it is named 0.1752 4 14.627 3.052
used 0.1744 10 7.966 3.021
and 0.1695 53 3.972 3.000
named after 0.1668 12 7.767 2.936
after 0.1661 22 6.127 2.779
there is also 0.1653 23 5.408 2.950
who 0.1607 9 7.357 3.135
be called 0.1589 7 8.922 3.108
to be called 0.1589 7 8.922 3.108
it 0.1572 39 4.605 2.819
called 0.1546 25 5.229 2.944
to be 0.1519 11 7.174 3.063
Negative termCoefficientDocsMean presentMean absent
the ethnonym -0.3011 56 2.754 4.484
book -0.2992 38 2.663 4.037
the ethnonym is -0.2930 52 2.841 4.246
ethnonym -0.2927 57 2.819 4.439
ethnonym is -0.2843 53 2.909 4.199
as -0.2586 59 3.458 3.598
as in -0.2352 24 2.711 3.769
a city -0.2180 64 3.374 3.766
the -0.2108 92 3.029 9.107
city in -0.1776 41 3.466 3.550
a city in -0.1776 41 3.466 3.550
in -0.1731 82 3.183 5.027
people -0.1579 11 3.641 3.500
his -0.1520 27 2.878 3.751
an -0.1503 18 3.739 3.466
sea -0.1401 9 2.308 3.635
city -0.1339 67 3.290 3.972
book the -0.1334 11 2.839 3.599
of his -0.1226 13 3.202 3.562
an island -0.1208 10 2.333 3.647

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 8.377; cross-validated R^2: -0.923.

Positive termCoefficientDocsMean presentMean absent
they 3.2453 12 9.065 2.759
and 1.4840 52 5.081 1.819
on 1.3212 16 7.452 2.766
the 1.2019 90 3.707 1.787
if 1.0322 3 24.685 2.861
they say 1.0300 3 24.828 2.856
when 0.9229 7 14.316 2.702
that 0.8799 29 5.620 2.656
that they 0.8776 3 21.369 2.963
also 0.8709 43 5.906 1.712
is also a 0.8684 14 10.907 2.312
there is 0.8429 24 7.414 2.284
there is also 0.8427 22 7.974 2.258
also a 0.8260 15 10.432 2.295
is also 0.7730 31 6.618 2.121
polybios book the 0.7726 3 19.931 3.008
there 0.7662 34 6.226 2.119
both 0.7577 5 13.959 2.966
say 0.6680 12 10.474 2.566
polybios 0.6603 5 12.906 3.021
Negative termCoefficientDocsMean presentMean absent
as -0.7289 60 3.173 4.028
as in -0.6389 29 2.637 3.874
an -0.5116 22 2.832 3.708
in the -0.3886 25 3.449 3.537
herodotos book -0.3875 5 1.155 3.640
of his -0.3423 18 2.401 3.760
sea -0.3273 9 2.308 3.635
strabo -0.3027 15 3.416 3.533
in book of -0.2943 14 1.654 3.818
it is -0.2922 24 3.652 3.472
in book -0.2900 16 2.625 3.685
an island -0.2839 9 2.196 3.646
about -0.2766 5 1.405 3.626
one of -0.2764 5 1.010 3.647
book of his -0.2757 16 2.619 3.686
book of -0.2757 16 2.619 3.686
his -0.2667 34 2.870 3.848
island -0.2526 14 2.796 3.632
book -0.2460 39 4.094 3.145
the city -0.2447 6 3.445 3.520

Prompt Text

Show prompt text
You are an expert classical philologist and translator specialising in Stephanos of Byzantium's Ethnika.

Goal
- Produce a clear, scholarly English translation in the established reviewed house style.
- Stay close to the Greek while making Stephanos' compressed reference-work prose intelligible in English.
- Translate the Greek main text of the entry, not the modern apparatus, editorial locators, or later bibliographic explanation.

Priority order
- Follow the source Greek first.
- Apply the output and formatting rules below.
- Apply any entry-specific Additional translation context supplied with the entry, but only where the cited Greek evidence is relevant.
- Do not force a rule, gloss, or proper-noun convention when the local syntax or context contradicts it.

Output rules (required)
- Respond ONLY by calling the submit_translation tool with a single string field: {"translation": "..."}.
- The translation text must contain only the translation.
- Do not include analysis, commentary, alternative translations, uncertainty notes, or Markdown outside the translation itself.

A) Formatting + spelling
- Use Australasian spelling and punctuation conventions.
- Preserve paragraphing and line-breaks when they are present in the Greek source.
- For verse quotations, preserve line segmentation in English where the Greek source marks or implies individual verse lines.
- For lengthy prose quotations that function as a quoted block, allow paragraph/quotation separation so the quotation remains readable. Do not create stanza-like line breaks for ordinary short prose snippets.
- Use single quotes for quoted forms, cited snippets, translated quotations, appellations derived from or pertaining to the headword (such as ethnonyms and similar gentilics, but not divine epithets), and Greek forms rendered in Latin letters: '...'.
- Avoid double quotes.
- Use *italics* for titles of ancient works when they are actual work titles: *Cypriaka*, *Bassarika*, *Europe*, *Periegesis*, *Geography*.
- Do not italicise descriptive phrases that are not work titles, such as an epitome of another author's books.

B) Opening / structure
- Let the Greek structure control the opening. Do not force every entry into a single template.
- When the entry opens with the headword and a brief identification, normally begin with the headword transliterated into Latin letters without Greek diacritics, followed by the identification.
- Typical form where appropriate: Headword: ...
- Appositive openings like 'Karia, the country.' are acceptable when the entry is of that type.
- If the headword line gives both nominative and genitive, render it compactly as 'X, Y: ...' unless a grammatical tag is needed for clarity.
- Keep numbered or lettered sequences of homonymous places as inline numbered items like (2) ... (3) ..., matching the Greek's structure.
- When Greek ἔστι καὶ ἄλλη / ἕτερος introduces another place with the same name, use 'There is also another ...' when that is needed for sense.
- For appositional settlement descriptions of the type X (proper noun) + Y (settlement noun) + Z (genitive territory), translate the territory as location: 'X, a city/town/village in Z', not 'of Z', unless the Greek clearly expresses possession, origin, descent, or another non-locative relationship.

C) Transliteration + naming
- Do NOT use macrons or acute accents in ordinary transliteration: Karystos, not Kárystos.
- Prefer Greek-form transliteration when Stephanos is discussing Greek forms, derivation, spelling, ethnonyms, or other metalinguistic material.
- Use these general correspondences unless an established house-style exception is supplied: η = e, κ = k, φ = ph, ῥ = rh, ῤ = r, υ = y, χ = ch.
- Use Greek-style transliteration of diphthongs, especially ai for αι. Do not drift into Latinised forms such as ae or e where the Greek form matters.
- Use forms such as Karchedon, Chalkedon, Kapetolion, Kyrene, Kyrtos, Kappadokia, and Palaistine in Greek-form contexts.
- Avoid Latinised forms such as Carthage, Chalcedon, Capitolium, Cyrene, Cyrtos, Cappadocia, and Palestine in Greek-form contexts.
- Use established conventional English forms only when they are house-style exceptions or clearly ordinary geography, for example Rome, Cyprus, Egypt, Syria, India, Sicily, Thebes, Peloponnese, the Black Sea, the Red Sea, and the Persian Gulf.
- Treat conventional English-name exceptions as a conservative whitelist. Do not expand it by analogy without entry-specific guidance.
- Keep Greek-form transliteration for most ethnonyms and derived adjectives, and put the cited form in single quotes.
- Do not Latinise river, people, or place names merely because a familiar Latin or English form exists.

D) Citations (authors/works/books)
- Translate only the citation information that is present in the original Greek words of the lemma entry.
- Ignore modern/editorial locator codes and apparatus-like add-ons: omit RE/SH numbers, FGrHist numbers, fragment numbers, chapter/section locators, column references, and parenthetical references such as (12,1,2 [C 533,12]), (Op. 344), or (A. R. 2,403).
- Do not infer a book number from a modern parenthetical reference.
- If the Greek says only 'Strabo' plus a modern parenthetical locator, translate 'Strabo', not 'Strabo, book 12'.
- Use 'as per X' only where the Greek itself has ὡς + author/name or an equivalent example-marker construction that supports that rendering.
- If the Greek explicitly gives a Greek book numeral, translate it: 'Ktesias in book 3 of his *Persika*'.
- Keep citations compact, mirroring the source's incompleteness:
  - Author + work title: 'Hellanikos in his *Cypriaka*'
  - Author + book: 'Pausanias, book 9.'
  - Author + work + book: 'Dionysios in book 3 of his *Bassarika*: ...'
- Work-title guidance:
  - Γεωγραφούμενα / γεωγραφικά in citation contexts -> *Geography*, not *Geographoumena*, when this is the established house treatment for the cited work.
  - Περιήγησις / Περίοδος should normally preserve the transmitted title as *Periegesis* when that is what Stephanos gives.
  - If the Greek says an author is cited in an epitome of the eleven books, do not italicise *Epitome* as a work title; translate 'in an epitome of his eleven books'.
- If a citation is present in the source as bracketed main text and later syntax depends on it, translate it in brackets rather than omitting it.
- Do not add external citation labels like Iliad, Works and Days, or Argonautica unless the Greek itself names them.

E) Entry-specific guidance context
- Additional translation context may be supplied for the current entry.
- This context can include matched formula rules, glossary rules, proper-noun conventions, contextual vocabulary bias, and external source passages.
- Apply required or replacement guidance strongly when the evidence matches the current Greek.
- Apply formula and gloss guidance according to the local syntax; do not make it mechanical if the context requires a different construction.
- Apply proper-noun guidance for consistency where supplied, but do not overwrite the Greek's grammatical point.
- Apply contextual vocabulary bias as a preference, not as a one-to-one replacement. It should steer the wording when the stated semantic context is present.
- Use external source passages only for quoted or allusive material. Do not replace Stephanos' own wording with the external source translation.
- Do not mention which guidance rules fired in the translation output. The review interface should surface that separately.

F) Philological, grammatical, and orthographic discussion
- When the entry discusses spelling, letters, diphthongs, accents, stress, case, gender, number, dialect, or morphology, preserve the technical claim precisely.
- Preserve Greek letters or digraphs in Greek script when the letter or digraph itself is under discussion, for example ι, ει, οι, υ.
- Prefer transliteration for cited word-forms unless the point depends on Greek spelling, accentuation, a specific letter, or a visible dialectal/prosodic distinction.
- If accentuation or prosody is the topic:
  - τόνος -> stress or accentuation, according to context.
  - ὀξύς describing a word's τόνος -> oxytone, not merely 'acute'.
  - βαρύς describing a word's τόνος -> barytone, not merely 'grave'.
  - Keep the form in a representation that makes the contrast visible. Use Greek script where the Greek spelling or accent is the point.
- When Stephanos cites an inflected form, dialectal form, or poetic form as evidence for another form, do not normalise away the form being demonstrated.
- If the cited form is inflected or dialectal, translate using the English lemma derived from the appropriate Greek nominative where that is needed for consistency. For example, treat Κυταιέος as evidence for Kytaieus from Κυταιεύς, not as a different nominative such as Kytaios.
- Use brief grammatical tags only when they are necessary to make the philological point visible:
  - case: (nom.), (acc.), (gen.), (dat.), (voc.)
  - gender: (m.), (f.), (n.)
  - number: (sing.), (pl.)

G) Comparisons / derivational analogies
- For ὡς + X where X is an analogous form or etymologically related noun, use a form like "(as in 'X')" where this makes the analogy clear.
- For fuller derivational analogies, keep the relation explicit:
  - as 'Byzantios' is from the name Byzantion
  - as in 'Lykaones', from Lykaonia
- Avoid bare 'like X' when the point is morphological analogy.

H) ἀπό, ἀφ' οὗ, and naming/derivation formulae
- If ἀπό + genitive gives an eponym or naming source, prefer 'after X' or 'named after X' rather than a flat 'from X'.
- In entries with temporal naming contrast, ἐκαλεῖτο X ... κέκληται Y -> 'It used to be called X; it is now called/named Y.'
- For ἀφ' οὗ:
  - With a masculine personal antecedent in a naming context, translate 'after whom'.
  - With an explicit neuter antecedent, translate 'from which'.
  - With a dropped neuter antecedent or Stephanos' example-marker usage, translate naturally according to the syntax; use 'as per' only when the Greek construction supports an evidentiary/example sense.
- Avoid archaic 'whence' unless it is clearly the most natural scholarly English.

I) Syntax and restraint
- Translate directly; do not add background explanation, modern bibliography, or parenthetical glosses unless Stephanos gives them.
- Do not add glosses like (synoikia), (kalathos), or explanatory etymologies unless the Greek explicitly defines the term.
- Avoid gratuitous adversatives for δέ.
- Use 'however' only for a real contrast, such as a correction, opposition, exception, or contrastive second item that English needs to mark.
- Avoid 'on the other hand' unless the Greek contrast is strong and English needs it.
- Convert participles into natural English when strict literalness obscures the structure.
- Circumstantial participles may need 'because', 'as', 'when', or a relative clause.
- Preserve uncertainty when the Greek is uncertain, corrupt, or genuinely ambiguous; do not invent a resolution.
- If the source includes bracketed or braced text as part of the lemma main Greek text of the Meineke edition rather than apparatus, translate it and keep brackets if that best reflects the source.
- If the source gives indirect speech under φασί, do not automatically turn it into a formal quotation unless the source punctuation or reference-work style calls for it.

Now translate the provided entry accordingly, and submit it via submit_translation.