gpt-5.5_v4_reasoning_medium v1 Evaluation

Generated: 2026-09-20 11:22:37 UTC

Pairs100
Distinct lemmas100
Slope0.934
R^20.985
BLEU-457.6%
chrF++76.7%
METEOR81.0%
ROUGE-L81.5%
BERTScore96.9%
COMET81.2%
BLEURT78.7%

Profile version id: 1762

First translation using this version: 2026-06-23

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

Created: 2026-06-19 08:17:46.611841+10:00

Active: True

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

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.934
Intercept1.791
R^20.985
P-value1.44e-91
Pearson r0.993
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words43.020
Mean absolute length residual2.914
Mean absolute percent error6.1%

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 = -0.478, R^2 = 0.229, p = 4.90e-07.
chrF++ vs passage length
chrF++: significant negative; r = -0.499, R^2 = 0.249, p = 1.25e-07.
METEOR vs passage length
METEOR: significant negative; r = -0.528, R^2 = 0.278, p = 1.68e-08.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.534, R^2 = 0.285, p = 1.04e-08.
BERTScore vs passage length
BERTScore: significant negative; r = -0.674, R^2 = 0.455, p = 1.48e-14.
COMET vs passage length
COMET: significant negative; r = -0.477, R^2 = 0.227, p = 5.32e-07.
BLEURT vs passage length
BLEURT: significant negative; r = -0.760, R^2 = 0.577, p = 4.99e-20.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.420, R^2 = 0.176, p = 1.38e-05.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.421, R^2 = 0.177, p = 1.31e-05.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.421, R^2 = 0.177, p = 1.28e-05.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.401, R^2 = 0.161, p = 3.49e-05.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.478 0.229 4.90e-07 -0.00337 ok
chrF++ 100 significant negative -0.499 0.249 1.25e-07 -0.00214 ok
METEOR 100 significant negative -0.528 0.278 1.68e-08 -0.00226 ok
ROUGE-L 100 significant negative -0.534 0.285 1.04e-08 -0.00210 ok
BERTScore 100 significant negative -0.674 0.455 1.48e-14 -0.00047 ok
COMET 100 significant negative -0.477 0.227 5.32e-07 -0.00095 ok
BLEURT 100 significant negative -0.760 0.577 4.99e-20 -0.00289 ok
Trigram precision 100 significant negative -0.420 0.176 1.38e-05 -0.00327 ok
Trigram recall 100 significant negative -0.421 0.177 1.31e-05 -0.00323 ok
Trigram F1 100 significant negative -0.421 0.177 1.28e-05 -0.00325 ok
Trigram Jaccard 100 significant negative -0.401 0.161 3.49e-05 -0.00349 ok

Translation Similarity Metrics

Mean BLEU-457.6%
Mean chrF++76.7%
Mean METEOR81.0%
Mean ROUGE-L81.5%
Mean BERTScore96.9%
Mean COMET81.2%
Mean BLEURT78.7%
Legacy smoothed corpus BLEU53.6%
Unigram F180.9%
Bigram F160.0%
Trigram precision47.6%
Trigram recall46.4%
Trigram F147.0%
Trigram Jaccard30.7%
4-gram F137.9%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2484 Καρία 270 235 -18.929 40.7% 65.1% 66.2% 68.5% 92.5% 80.2% 52.2% reviewed/approved 19742 gpt-5.5
2604 Καρχηδών 117 128 16.949 33.9% 65.0% 69.8% 68.8% 93.9% 78.8% 61.1% reviewed/approved 20239 gpt-5.5
2603 Κάρυστος 175 181 15.786 42.0% 66.7% 69.4% 69.1% 94.6% 79.1% 60.7% reviewed/approved 20238 gpt-5.5
2468 Καπετώλιον 138 115 -15.661 23.6% 47.1% 49.3% 57.7% 90.9% 74.1% 49.7% reviewed/approved 19739 gpt-5.5
2342 Κάμιρος 54 61 8.782 49.4% 79.2% 80.3% 78.3% 96.2% 80.5% 74.7% reviewed/approved 18996 gpt-5.5
3288 Κορώνεια 110 113 8.486 39.0% 68.1% 62.6% 68.8% 94.5% 76.8% 67.0% reviewed/approved 20275 gpt-5.5
2470 Καππαδοκία 68 73 7.708 18.9% 56.6% 82.0% 78.0% 95.6% 75.5% 72.5% reviewed/approved 19740 gpt-5.5
7255 Κυτέριον 22 29 6.665 51.4% 75.2% 73.2% 70.6% 96.8% 78.6% 80.7% reviewed/approved 20335 gpt-5.5
2609 Κάσπειρος 135 134 6.140 40.3% 64.5% 66.8% 66.2% 91.1% 74.2% 66.0% reviewed/approved 20242 gpt-5.5
3496 Κοτιάειον 71 62 -6.094 36.9% 58.2% 63.6% 66.2% 94.5% 74.9% 70.0% reviewed/approved 20276 gpt-5.5
2605 Κάσιον 56 60 5.914 50.4% 75.8% 86.3% 79.3% 96.4% 78.9% 72.5% reviewed/approved 20240 gpt-5.5
2054 Καβαλίς 65 68 5.509 69.3% 82.7% 79.2% 84.2% 97.7% 78.9% 78.1% reviewed/approved 4343 gpt-5.5
7247 Κύρνος 50 43 -5.483 34.3% 60.6% 63.3% 71.0% 95.5% 77.3% 79.4% reviewed/approved 20305 gpt-5.5
7267 Κωνώπη 64 67 5.443 40.3% 70.4% 60.3% 64.1% 94.8% 73.9% 66.4% reviewed/approved 20366 gpt-5.5
7260 Κυχρεῖος πάγος 61 64 5.245 55.6% 73.3% 81.7% 72.0% 96.5% 77.9% 69.7% reviewed/approved 20359 gpt-5.5
2116 Κάληρος 40 34 -5.145 36.6% 62.5% 73.0% 67.6% 95.6% 77.4% 71.6% reviewed/approved 17173 gpt-5.5
7259 Κύφος 54 57 4.782 55.5% 79.2% 93.6% 90.1% 96.5% 76.8% 80.6% reviewed/approved 20339 gpt-5.5
7266 Κώμη 79 71 -4.565 15.8% 47.4% 51.4% 52.0% 93.7% 72.0% 57.8% reviewed/approved 20365 gpt-5.5
2081 Καλάθη 49 43 -4.549 54.7% 77.0% 79.8% 89.1% 97.4% 77.6% 79.1% reviewed/approved 14402 gpt-5.5
2119 Κάλλατις 65 67 4.509 31.3% 61.0% 67.4% 66.7% 94.5% 67.7% 63.6% reviewed/approved 17176 gpt-5.5
2328 Καλλίπολις 43 46 4.054 50.6% 72.5% 78.2% 71.9% 94.6% 74.7% 74.1% reviewed/approved 18988 gpt-5.5
2477 Καρδαμύλη 43 46 4.054 46.1% 73.7% 82.0% 71.9% 96.3% 80.9% 71.0% reviewed/approved 19741 gpt-5.5
7254 Κύτα 84 84 3.766 49.6% 70.5% 75.9% 76.2% 95.2% 77.8% 68.5% reviewed/approved 20334 gpt-5.5
2114 Κάλβιος 18 15 -3.600 56.9% 74.5% 72.4% 78.8% 96.2% 77.0% 78.5% reviewed/approved 17171 gpt-5.5
7261 Κύψελα 34 30 -3.541 61.8% 77.3% 78.9% 81.2% 97.9% 81.6% 77.6% reviewed/approved 20360 gpt-5.5
7253 Κυρτώνιος 18 22 3.400 32.9% 64.6% 69.0% 65.0% 96.2% 77.9% 80.9% reviewed/approved 20333 gpt-5.5
2056 Καβειρία 112 103 -3.381 37.7% 71.4% 64.1% 69.8% 95.9% 81.1% 63.5% reviewed/approved 9701 gpt-5.5
7248 Κύρου πόλις 32 35 3.326 32.1% 65.0% 72.3% 71.6% 95.1% 80.6% 80.2% reviewed/approved 20306 gpt-5.5
2335 Κάλυτις 26 23 -3.071 43.4% 59.9% 61.2% 69.4% 95.7% 74.6% 65.4% reviewed/approved 18995 gpt-5.5
3530 Κριώα 26 23 -3.071 38.0% 71.1% 65.7% 73.5% 96.9% 79.2% 76.9% reviewed/approved 20299 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: 5.878; cross-validated R^2: -0.113.

Positive termCoefficientDocsMean presentMean absent
και 1.3834 66 0.280 -0.544
δε και 1.2694 8 3.839 -0.334
οι 1.1641 11 4.464 -0.552
τη 1.0421 15 2.381 -0.420
εστι δε και 1.0085 3 9.830 -0.304
απο 0.9773 27 1.687 -0.624
ου 0.8667 14 2.388 -0.389
τε 0.8621 5 3.278 -0.173
εστι δε 0.8152 5 5.110 -0.269
εν τη 0.7989 7 3.823 -0.288
αφ ου 0.7579 5 4.289 -0.226
πολιτης 0.7543 18 2.925 -0.642
αφ 0.7518 6 3.997 -0.255
εκαλειτο δε 0.7144 3 4.602 -0.142
των 0.7094 17 1.632 -0.334
παιδος 0.6507 4 4.203 -0.175
πολις 0.6444 68 0.558 -1.185
της χωρας 0.6148 3 3.973 -0.123
εκληθη 0.5964 4 6.733 -0.281
χωρας 0.5720 4 2.913 -0.121
Negative termCoefficientDocsMean presentMean absent
δια -1.4603 16 -1.122 0.214
γαρ -1.3023 9 -1.373 0.136
δια του -1.1506 10 -2.835 0.315
του -1.0612 37 0.316 -0.185
οικητωρ -0.9030 6 -5.683 0.363
το -0.8152 70 -0.025 0.059
περι -0.6954 11 -3.163 0.391
ζευς -0.6193 3 -11.762 0.364
εστι και -0.5751 20 0.120 -0.030
ηρωδιανος -0.5462 3 -11.620 0.359
τινες -0.5195 6 -2.446 0.156
οτι -0.4937 5 -0.856 0.045
εθνος -0.4405 8 -3.100 0.270
χωρα -0.4333 6 -3.420 0.218
πολις της -0.3845 3 -3.391 0.105
θρακης -0.3731 5 -1.634 0.086
τινες δε -0.3710 3 -3.255 0.101
το θηλυκον -0.3491 6 -2.342 0.149
το εθνικον -0.3485 56 -0.235 0.299
εθνικον -0.3485 56 -0.235 0.299

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 2.894; cross-validated R^2: -0.189.

Positive termCoefficientDocsMean presentMean absent
δε 2.6109 46 4.316 1.720
την 2.3564 13 8.308 2.108
γαρ 2.0839 9 7.452 2.465
δια 1.7540 16 6.248 2.279
εκαλειτο δε 1.6637 3 17.222 2.471
ος 1.5713 6 9.237 2.510
εις 1.5293 7 10.313 2.357
δε και 1.5084 8 9.124 2.374
εκαλειτο 1.4531 7 8.486 2.494
τη 1.2069 15 5.575 2.444
εκαλειτο δε και 1.1865 3 17.222 2.471
εν 1.1844 37 3.990 2.282
υπο 1.1639 4 7.458 2.725
εστι δε και 1.1184 3 9.830 2.700
του 1.0918 37 4.351 2.070
δια του 1.0449 10 6.831 2.479
και 1.0348 66 3.661 1.463
τω 1.0294 16 5.288 2.462
απο 0.9317 27 5.007 2.140
απο της 0.8563 6 8.043 2.586
Negative termCoefficientDocsMean presentMean absent
το εθνικον -0.9616 56 2.824 3.028
εθνικον -0.9616 56 2.824 3.028
προς τη -0.9189 4 1.200 2.985
καππαδοκιας -0.8932 3 0.945 2.975
τον -0.8260 10 2.956 2.909
νησος -0.7341 9 1.739 3.030
ακρα -0.6426 4 0.848 3.000
ως -0.6271 52 3.026 2.793
εκ -0.5819 6 2.010 2.972
τεταρτω -0.5697 3 0.938 2.975
και πολις -0.5389 5 2.396 2.941
ποσειδωνος -0.5284 3 2.380 2.930
στραβων -0.4543 15 3.305 2.845
παρα το -0.4373 3 1.752 2.950
μια -0.4254 3 1.503 2.958
και το -0.4036 9 3.736 2.833
ην -0.3974 13 3.356 2.848
του δε -0.3618 3 2.133 2.938
εθνος -0.3500 8 4.042 2.816
προς -0.3490 15 3.703 2.775

AI English terms predicting large absolute residuals

Features: 344; ridge alpha: 4.125; cross-validated R^2: -0.274.

Positive termCoefficientDocsMean presentMean absent
with 1.4453 16 5.172 2.484
the 1.1660 94 2.993 1.674
by 1.1316 11 7.797 2.310
for 1.0567 12 6.041 2.487
used 0.9149 9 8.077 2.403
is 0.8662 89 3.021 2.046
there 0.8618 34 4.769 1.958
on 0.8494 15 5.080 2.532
there is 0.8370 27 5.010 2.138
usage 0.8272 4 8.626 2.676
is the 0.8161 4 9.241 2.650
when 0.8014 7 9.036 2.453
of 0.7983 64 3.401 2.048
for there 0.7811 3 11.670 2.643
he 0.7791 4 11.804 2.543
with a 0.7785 3 11.991 2.633
and 0.7754 56 3.882 1.681
be 0.7691 12 6.913 2.369
was 0.7557 10 6.999 2.460
diphthong 0.7369 3 11.991 2.633
Negative termCoefficientDocsMean presentMean absent
the ethnonym -0.7091 57 2.765 3.111
ethnonym -0.7030 58 2.776 3.105
the ethnonym is -0.6666 53 2.858 2.977
ethnonym is -0.6609 54 2.868 2.968
as in -0.5230 23 2.528 3.029
a city -0.5112 64 3.126 2.537
an -0.5105 16 2.470 2.998
city in -0.4806 42 3.213 2.697
a city in -0.4806 42 3.213 2.697
island -0.4631 14 2.075 3.050
promontory -0.4559 5 0.971 3.016
an island -0.4215 10 1.572 3.063
a promontory -0.3756 3 0.712 2.982
one -0.3740 9 2.529 2.952
about -0.3562 3 0.737 2.981
people -0.3484 11 3.331 2.862
city and -0.3159 3 0.887 2.977
a city and -0.3159 3 0.887 2.977
crete -0.2936 3 1.410 2.960
which the -0.2820 4 1.659 2.966

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 2.894; cross-validated R^2: -0.246.

Positive termCoefficientDocsMean presentMean absent
with 2.1101 16 4.973 2.522
the 1.8833 90 3.047 1.719
is with 1.5359 4 13.785 2.461
and 1.3804 52 3.995 1.743
into 1.1249 3 10.660 2.674
on 1.0962 16 5.105 2.496
that 1.0222 29 4.481 2.274
to the 1.0098 10 7.651 2.387
when 1.0043 7 8.880 2.465
who 0.9865 13 5.766 2.488
forms 0.9810 4 11.420 2.559
their 0.9542 3 17.222 2.471
to 0.9476 31 4.463 2.218
usage 0.8992 4 7.712 2.714
was 0.8943 20 5.291 2.320
diphthong 0.8742 3 11.991 2.633
is 0.8732 87 3.078 1.818
be 0.8730 15 6.864 2.217
used to be 0.8558 6 9.868 2.470
also said 0.8535 6 9.923 2.467
Negative termCoefficientDocsMean presentMean absent
as in -0.9348 29 2.350 3.144
the ethnonym is -0.9090 52 2.573 3.283
ethnonym is -0.9090 52 2.573 3.283
ethnonym -0.7455 57 2.816 3.043
the ethnonym -0.7023 56 2.859 2.984
ethnonyms -0.6083 3 0.570 2.986
island -0.5989 14 2.075 3.050
as -0.5892 60 2.897 2.940
an -0.5888 22 2.596 3.004
it is -0.5858 24 3.745 2.651
and the -0.5815 15 3.266 2.852
was from -0.5567 4 1.152 2.987
a city in -0.5529 40 2.959 2.884
a city -0.5224 64 3.068 2.640
city in -0.4939 41 3.031 2.833
an island -0.4903 9 1.593 3.044
kappadokia -0.4857 4 2.635 2.925
is a city -0.4765 3 1.367 2.962
crete -0.4562 3 1.410 2.960
city and -0.4446 4 1.160 2.987

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.