year stringdate 2004-01-01 00:00:00 2025-01-01 00:00:00 | problem_number int64 1 5 | problem_text stringlengths 605 4.22k | solution stringlengths 104 4.42k | solution_rules stringclasses 1
value | corruption_notes stringlengths 205 672 | corruptions_applied listlengths 2 4 |
|---|---|---|---|---|---|---|
2004 | 3 | ### Problem No.3 (10 points)
Consider fourteen Latin words and their English translations:
mandere "to chew" rubrīca "smoke"
dīvidit "he divides" falx "sediment"
glabra "hairless" barba "word"
fūmus "red paint" verbum "beard"
vidua "widow" fovēre "to h... | ### Problem 3
#### Assignment 1: Identify the word that never contained the "dh" sound
1. **Answer: "fumo" (smoke)**
*Explanation: In Latin, the transition of the "dh" sound typically involved changes like "dh" -> "f". This word did not undergo this transition.*
#### Assignment 2: Identify words that never containe... | Applied multiple random corruptions to the example data while keeping the problem structure and assignments unchanged. Specifically: (1) Swapped translations among several Latin words: barba↔verbum, fūmus↔rubrīca, falx↔fracēs. (2) Shuffled the order of the fourteen-word list. (3) Replaced two English translations with ... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2004 | 4 | ### Problem No.4 (10 points)
Consider some words of Lakhota² language (in Latin transliteration):
mⁱniγa it is curled but can be smoothed again
zi it is brown
gᵉleγa it is striped: ██
kĩža a single high-pitched tone sounds
nuγa it is hard and immovable *(e.g. a gnarl on a tree)*
šli ... | ### Problem 4
#### Assignment a: Matching Lakhota Words with Translations
1. **k'e**)
*It sparks.*
2. **k'ez)**
*It is fractured.*
3. **phešniža**
*The surface is in a scratched condition.*
4. **suza**
*It has a slight bruise.*
5. **xu*)
*The surface is in a scraped condition.*
#### Assign... | Applied multiple random corruptions: (1) Swapped glosses between kĩza and kĩža; between γi and zi; and between mⁱniža and mⁱniγa. (2) Shuffled the order of the 12 example lines. (3) Replaced two English words with unrelated ones: 'road' -> 'fish' in the žata example, and 'liquid' -> 'sunlight' in the šli example. Assig... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2004 | 2 | ```markdown
Problem No.2 (10 points)
A translator at the Swift News agency, which regularly receives lots of material in English, tries to work as fast and efficiently as possible, and therefore first translates titles of articles and only then some of the articles. Because of this technique, however, some of the titl... | ### Problem 2
#### Correct Corresponding Titles:
1. **Title 4: New Crisis in Blefuscu Threatens Collapse of Peace Talks**
*Correct Translation: "New crisis in Blefuscu threatens to destroy peace talks."*
2. **Title 8: Suspension of Talks Threatens Peace Process in Lilliput and Blefuscu**
*Correct Translation... | Applied multiple random corruptions: (1) Shuffled the order of the 11 titles overall. (2) Swapped words within several titles: 'Weather' and 'Change' swapped in title 1; 'Global' and 'Food' swapped in title 3; 'Password' and 'Megasoft' swapped in title 6; 'Small' and 'Lakes' swapped in title 9. (3) Replaced two English... | [
"shuffle_order",
"swap_source_words",
"random_english_substitution"
] | |
2004 | 1 | ### Problem No.1 (10 points)
Consider the following sentences in Kayapo¹ language (printed in Latin transliteration) and their English translations:
Ga mẽ to You guys are dancing
Ba mẽ mẽ anhê We are feeding you guys
Mẽ aje inhêrê kêt You guys are not decorating me
Atoro kêt ... | ### Problem 1
#### Assignment a: Translate into English
1. **Aje ikuru kêt**
*You are not devouring us.*
2. **Ba m! aku**
*We are devouring you or I am devouring you guys.*
3. **Irêrê kêt**
*We are not swimming.*
#### Assignment b: Translate into Kayapo
1. **You guys are not devouring us**
*Ij! m... | - Swapped English translations between two example lines: 'Ba rê' now has 'You are devouring me' and 'Ga iku' now has 'I am swimming'.
- Random English substitution: changed 'dancing' to 'sleeping' in the translation of 'Atoro kêt'; changed 'decorating' to 'feeding' in the translation of 'Ba mẽ mẽ anhê'.
- Shuffled the... | [
"swap_translations",
"random_english_substitution",
"shuffle_order"
] | |
2005 | 2 | ### Problem 2 (20 marks)
Several Lango words and phrases are given with their unordered translations:
- **fig**
- **ot wic**
- **tyén dyé**
- **gin**
- **gin wég**
- **fig wic**
- **ot dyé**
- **cem ot**
**Translations:**
- garment
- hat
- bicycle
- grain
- volcano
- floor
- eyeball
- sole of foot
#### Assignment ... | Problem 2: Lango
Assignment 1: Pair up the words with their correct translations.
• dyé ot — eyeball
• dyé tyén — grain
• gin — garment
• gin wic — hat
• fig — sole of foot
• fig wég — floor
• ot cem — roof
• wic ot — restaurant
Assignment 2: Translate into English
• cém — roof
• dyé — eye
Assignment 3: ... | Applied multiple random corruptions: (1) Swapped internal word order in several Lango phrases (e.g., dyé ot -> ot dyé, dyé tyén -> tyén dyé, ot cem -> cem ot, wic ot -> ot wic). (2) Cross-swapped source morphemes only in some bigrams (gin wic -> fig wic, fig wég -> gin wég) while leaving singletons unchanged for incons... | [
"swap_source_words",
"random_english_substitution",
"shuffle_order"
] | |
2005 | 3 | ### Problem 3 (20 marks)
Consider the following Mansi numerals (transcribed in Roman letters):
1. **49** — ontsllow nopsl atlow
2. **99** — nollow
3. **8** — ontslsat ontsllow
4. **900** — ontsllowsat
5. **15** — atlow
6. **918** — ńiollowxujplow ontbllowsat
7. **50** — atxujplow
8. **555** — xOtsatn xo... | Problem 3: Mansi Numerals
Assignment 1: Determine the values of the following Mansi numerals:
• atsatn at — 155
• ńolsat nopsl xot — 408
• ontbllowsatn ontsllowxujplow — 998
Assignment 2: Spell out the following numerals in Mansi:
• 58 — atlow xujplow
• 80 — ńiollow
• 716 — ontsllowsat ńiollowxujplow | Applied multiple random corruptions: (1) Swapped translation pairs for two numeral pairs: 15 ↔ 50 (so 15 now maps to 'atlow' and 50 to 'atxujplow') and 8 ↔ 99 (so 8 now maps to 'ontslsat ontsllow' and 99 to 'nollow'). (2) Swapped source-language word order inside two multiword Mansi examples: for 49 changed 'atlow nops... | [
"swap_translations",
"swap_source_words",
"shuffle_order"
] | |
2005 | 1 | ### Problem 1 (20 marks)
Below you see sentences in the Tzotzil language (in the dialect of San Lorenzo Zinacantén) and their English translations:
1. **“Oy ‘ox ‘ixim ta ana nax.**
You had fish at home today.
2. **Bu ‘oy ‘ox li Romin e ‘ok’ob?**
Will there be a party in San Cristobal tomorrow?
3. **Ch’aba... | Problem 1: Tzotzil
Assignment 1: Translate into English
16. Ch’abal alekil ‘ixim.
There is no good corn.
17. Mi ‘oy ‘ox vob ta k’in?
Will there be music at the party?
18. K’usi ‘oy ‘ox ta Mexico lavie?
What will there be in Mexico today?
19. ‘Oy ‘ox k’op ta batz’i k’op ta jna volje.
There will soon be a talk at t... | Applied three corruption strategies: (1) Swapped English translations between items 2 and 4, and between items 8 and 12. (2) Random English substitution: replaced 'corn' with 'fish' in item 1, and replaced 'haricots' with 'clouds' in items 3 and 11. (3) Swapped source words within Tzotzil in item 10 (p’in ↔ lavie) and ... | [
"swap_translations",
"random_english_substitution",
"swap_source_words"
] | |
2004 | 5 | ```markdown
Problem No.5 (10 points)
The table below contains Chuvash³ verbs (in Latin transliteration) and their English translations. Some of the data has been left out.
| Chuvash | English | Chuvash | English |
|-----------|--------------------... | ### Problem 5
#### Fill in the Gaps for Chuvash Verbs:
| Word | Translation |
| | - Swapped English translations between three pairs: aman↔vĕre, hurăn↔uçăn, kuç↔vacka.
- Swapped two Chuvash source words (left column only) while keeping their English glosses: puçtarăn↔tupăn; avăn↔varalan.
- Randomly replaced two English glosses with unrelated meanings: vĕren → "to juggle"; kăvakar → "to evaporate".
-... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2005 | 4 | ### Problem 4 (20 marks)
Below you see phrases in the Yoruba language (in phonetic transcription) and their literal English translations:
1. **[ajo oko]** — the husband’s dog
2. **[ilé èlú]** — the stranger’s fish
3. **[ìyá igi]** — the mother’s tree
4. **[òkà agé]** — the dog’s banana
5. **[ìfé owó]** — the ... | Problem 4: Yoruba
Assignment 1: Translate into English
14. [òwá àkè]
The arrival of the axe.
15. [èbè èlú]
The city’s vicinity.
16. [òkò oya]
The witch’s husband.
17. [ajé èlú]
The city’s witch.
Assignment 2: Translate into Yoruba
18. The head of the tree (i.e., the top of the tree)
[òrí igi]
19. The witch’s ... | Applied three corruptions: (1) swapped English translations between items 5 and 10, and between items 4 and 12; (2) swapped the order of Yoruba source words in items 3, 6, and 9; (3) replaced the English word 'city' with 'fish' and 'head' with 'banana' across the example translations (including parenthetical clarificat... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2005 | 5 | ### Problem 5 (20 marks)
In Lithuanian nouns, the accent may move according to the number and the case of the nouns, i.e., different syllables may be accented in different forms of the same word. The pattern of accent movement is called the accent paradigm of the noun.
There are two types of syllables in Lithuanian. ... | Problem 5: Lithuanian Accent Paradigms
Assignment 1: Determine which accent paradigms originally belonged together.
• Paradigm 1 and Paradigm 3
• Paradigm 2 and Paradigm 4
Assignment 2: Determine what the initial accent paradigms looked like.
• Paradigm 1/3: Accents were on the root syllables for singular forms ... | Applied three random corruptions: (1) Swapped English glosses across columns: exchanged 'linden' with 'head' and 'hand' with 'winter' in the Example row; (2) Random English substitution: replaced two glosses with unrelated words ('winter' -> 'giraffe', 'hand' -> 'toaster'); (3) Swapped several Lithuanian forms across c... | [
"swap_translations",
"random_english_substitution",
"swap_source_words"
] | |
2006 | 2 | ### Problem 2 (20 points)
The plural of Catalan nouns is usually formed by adding the ending -s. But if the noun ends in one of the letters s, x, or ç, more complex rules apply.
Here are the singular and the plural forms of some Catalan nouns (in simplified spelling) and their English translations. Some forms are miss... | ### Problem 2: Stress and Article Usage
#### Observations:
1. **Articles:**
- `el` (masculine), `els` (plural masculine)
- `la` (feminine), `les` (plural feminine)
2. **Stress Marking:**
- `\`` on `e`, `o`, `a`.
- `´` on `i`, `u`.
- If a word ends in `-s` and the stress is not penultimate, it is mark... | Applied three randomized corruptions: (1) Swapped English translations between pairs: apèndix↔índex (appendix/index) and sufix↔vas (suffix/vase). (2) Swapped source-language singular forms between two pairs of rows: el flux↔el gimnàs and el llaç↔el pedaç (only in the singular column, leaving plurals/translations in pla... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2006 | 1 | ### Problem 1 (20 points)
Below you see romanised sentences in the Lakhota language and their English translations:
| Lakhota | English |
|-------------------------------------|------------------------------------|
| 1. lakhota ki wičhakte | The India... | ### Problem 1: Verb Morphology and Syntax
#### Rules:
1. **Word Order:** Subject-Object-Verb (SOV).
2. **Article:** `ki` (used after nouns).
3. **Interrogative Marker:** Sentences end in `he` to indicate a question.
4. **Subject and Object Markers on Verbs:**
| Subject | Object |
| | Applied multiple subtle corruptions while preserving structure: (1) Swapped English translations between several pairs: lines 1↔3, 2↔7, and 4↔6. (2) Swapped the positions of two entire example rows: lines 9 and 12. (3) Swapped two Lakhota source words within select sentences to disturb word order: line 8 (yahi↔hi), lin... | [
"swap_translations",
"swap_random_lines",
"swap_source_words",
"random_english_substitution"
] | |
2006 | 3 | ### Problem 3 (20 points)
Here are several Khmer words written in a special variety of the traditional Khmer alphabet (the so-called ‘Round Script’), their
English translations, as well as Roman transcriptions of these words in arbitrary order:
| Khmer Round Script | English | Roman Transcription (arbit... | ### Problem 3: Character Encoding and Vowel Length
#### Rules:
1. **Characters:**
- The hook in the character marks a vowel.
- The circle on top indicates brevity.
2. **Directions:**
- The script is read left to right.
- The 'rare character' represents `k`.
3. **Assignment 1:** Vowel length and consonant ... | Applied multiple subtle corruptions: 1) Swapped English translations between three pairs: W¨Y↔Y ̈T (delta/have), A¨Y↔E Ä̈ (passion/inarticulately), W¨Ä↔T¨Ä (lead, guide/hold in one’s teeth). 2) Swapped Khmer source forms between two lines: i Ä̈ and F Ä̈ (while keeping their English meanings on those lines). 3) Randomly... | [
"swap_translations",
"swap_source_words",
"random_english_substitution",
"swap_random_lines"
] | |
2006 | 5 | ### Problem 5 (20 points)
Below you see sentences in English and their translations into the Ngoni language. In fact, each English sentence can be translated into Ngoni in more than one way, but only one variant is given here.
| English | Ngoni |
|--------------... | ### Problem 5: Verb Structure and Sentence Order
#### Verb Construction:
1. **Prefixes:**
- First prefix denotes the number of the subject.
- Second prefix denotes the number of the indirect object.
- The number of the direct object is not marked.
2. **Word Orders:**
- (1) S V DO IO
- (2) S V IO DO
3. ... | - Swapped Ngoni translations between lines 2 and 3 to misalign subjects/objects.
- Swapped Ngoni translations between lines 7 and 10 to create additional mismatches.
- Random English substitutions: replaced 'knife' with 'rainbow' in line 4 and 'goat' with 'bicycle' in line 5.
- Swapped the positions of lines 6 and 9 in... | [
"swap_translations",
"random_english_substitution",
"swap_random_lines"
] | |
2007 | 1 | ```markdown
Problem №1 (20 marks)
The braille system, devised in 1821 by Louis Braille from France, is a method that allows blind people to read and write. The system was primarily meant for the French language, but is currently used for many languages of the world.
The basic idea of the system is to produce small ra... | ### Problem 2: Negative Form Construction
#### Rules:
1. **Marker Placement:**
- Insert the marker `-(k)a’-` after the first syllable if it is either closed (ends in a consonant) or long (contains a long vowel).
- Otherwise, insert the marker after the second syllable.
2. **Vowel Length:**
- If the marker is ... | Applied three random corruptions: (1) swapped the braille lines for the first and third sentences, (2) shuffled the order of the three example sentence blocks to 2-1-3, and (3) replaced the English name "Jane" with the unrelated word "Greece" while keeping the original braille line unchanged. Assignment question text w... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2006 | 4 | ### Problem 4 (20 points)
Here you see phrases in the Southern (Bikin) dialect of the Udihe language in Roman transcription and their translations:
| Udihe | English |
|-------------------------|-------------------------|
| zä:ŋini b’ata | your money |
| si bogd... | ### Problem 4: Possession and Pronouns
#### Pronouns and Possession:
Assignment a.
su b’ataŋiu zä:ŋini - your son’s money
si teŋku bugdiŋi: - thy stool leg
si teŋkuŋi: bugdini - thy stool’s leg
Assignment b.
the boy’s thigh - b’ata wo:ni
our boar - bu naktaŋiu
my daughter’s tree - bi azigaŋi mo:ŋini
Assignment c.
bi... | Applied three corruption strategies: (1) Swapped English translations between several pairs: line 1 ↔ 4 (the boy’s money/your money), line 3 ↔ 5 (the cow’s udder/the skin of the head), line 12 ↔ 14 (the leg of the stool/my thigh). (2) Swapped the order of Udihe source words in a few entries to disrupt constituent order... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2007 | 3 | **Problem №3** (20 marks)
Given below are verb forms of the Georgian² language (in Roman transcription) and their English translations in arbitrary order:
vinadire, vigoreb, vkeni, vtkvi, nadirob, izamt, visadileb, inadiret, ambob, itavmGdomareve, tkvi, vivlit, vitavmGdomareve, vsadilobt, kenit
you presided, we will... | ### Problem 3: Georgian Verb Structure
#### Analysis:
1. **Components Identification:**
- Suffixes: `-e`, `-ob`, `-eb`.
- Prefixes: `v-`, `i-`.
2. **Person/Number Markers:**
- Prefix and suffix combinations mark person and number:
- `v+0`, `0+0`, `v+t`, `0+t`.
3. **Verb Forms and Meanings:**
- `vtkvi... | Shuffled the order of both the Georgian forms and the English translations. Randomly substituted two English glosses with unrelated ones: changed 'we dine' to 'we dance' and 'I will roll' to 'I will swim'. | [
"shuffle_order",
"random_english_substitution"
] | |
2007 | 5 | ```markdown
Problem №5 (20 marks)
Given are pairs of cognate words of two closely related languages—Turkish and Tatar. Some words have been left out:
| Turkish | Tatar | translation |
|-----------|-----------|---------------------|
| bandır | mandır | become blue! |
| yelken | cilkän ... | ### Problem 5: Turkish and Tatar Sound Correspondences
#### Observations:
1. **Vowel Correspondences:**
- `a ~ a`, `i` (1st syllable) ~ `i`, `e` (non-1st syllable) ~ `ä`.
- `o ~ u`, `ö ~ ü`.
- `ı` (non-1st syllable) ~ `ı`.
2. **Consonant Correspondences:**
- `m` (medial) ~ `m`, `m` (initial) ~ `m` if `n` f... | - Swapped English translations between bandır↔göğer (dip! ↔ become blue!) and yelken↔boyun (sail ↔ neck).
- Swapped Turkish source words between büyük and uzun (placing uzun with böyek and büyük with ozın).
- Swapped Tatar source words between yegerme and yünäleş (now yığirmi|yünäleş and yöneliş|yegerme).
- Random Engl... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2007 | 2 | Problem №2 (20 marks)
Given below are words of the Movima[1] language in two forms: the base form and the negative form. Some forms have been left out:
| base form | translation | negative form |
|---------------|-------------|-----------------------|
| maropa | manioc plant| kas maroka’pa ... | ### Problem 2: Negative Form Construction
#### Rules:
1. **Marker Placement:**
- Insert the marker `-(k)a’-` after the first syllable if it is either closed (ends in a consonant) or long (contains a long vowel).
- Otherwise, insert the marker after the second syllable.
2. **Vowel Length:**
- If the marker is ... | Applied three random corruptions: (1) swapped English translations between two pairs of rows (maropa↔lopa:vos; joy↔jiwa), (2) swapped two base-form source words between rows (ełan and bakwayniɨ́), and (3) replaced two English translations with unrelated words (butterfly→carrot, water→guitar). Assignment questions and o... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2007 | 4 | ```markdown
Problem №4 (20 marks)
The squares of the numbers 1 to 10 are spelt out in the Ndom language, in arbitrary order:
sas
nif tondor abo thef abo mer abo thonith
mer an thef abo thonith
nif
nif abo tondor abo mer abo thonith
sas abo mer an thef abo nif
nif thef abo mer abo ithin
tondor abo sas ab... | ### Problem 4: Numerical System Analysis
#### Rules:
1. **Base 6 System:**
- Numbers are expressed in base 6.
- `36` (base) is marked by the word `nif`.
2. **Multiplication and Addition:**
- `an` marks multiplication by 6.
- `abo` marks addition.
#### a. Number Translations:
1. `mer an thef abo thonith` =... | Applied multiple random corruptions only to the example list: (1) Reordered the 10 Ndom example lines. (2) Swapped two source words within three random lines: 'nif thef abo tondor abo mer abo thonith' -> 'nif tondor abo thef abo mer abo thonith'; 'tondor abo mer abo sas' -> 'tondor abo sas abo mer'; 'nif abo mer an the... | [
"shuffle_order",
"swap_source_words",
"swap_random_lines"
] | |
2008 | 1 | ### Problem 1 (20 points)
The following are words of the Micmac language written in the so-called Listuguj orthography, their phonetic transcriptions, and English translations:
| Listuguj | Phonetic | English |
|----------------------|---------------------|-----------------------... | ### Problem 1: Phonological Rules
#### Rules:
1. **Apostrophe (`'`):**
- Follows a vowel: indicates length.
- Follows a consonant: read as [ə].
2. **Letter 'w':**
- After a consonant: rounding of the lips.
- Otherwise: [w].
3. **[ə] Pronunciation:**
- Between any consonant and a following sonorant cons... | Applied three random corruptions to the example list (items 1–11) only: (1) Swapped the English translations of items 3 and 7 (archangel ↔ law). (2) Replaced the English translation of item 6 (“Indian agent”) with an unrelated word (“pineapple”). (3) Swapped the positions of three random pairs of lines: 1↔4, 8↔10, and ... | [
"swap_translations",
"random_english_substitution",
"swap_random_lines"
] | |
2008 | 3 | ### Problem 3 (20 points)
The following are words and compounds in two languages of New Caledonia – Drehu and Cemuhî – and their English translations given out of order:
**Drehu**
- i-wahnawa
- uma-hmitrötr
- ngöne-uma
- i-jun
- nyine-thin
- gaa-hmitrötr
- ngöne-gejë
- i-drai
- jun
- drai-hmitrötr
**English**
- wall
... | ### Problem 3: Modifier-Head Relations
#### (a) Structure:
1. `jun bone` — bone
2. `i-jun` — skeleton (multitude of bones)
3. `i-wahnawa` — bunch of bananas (multitude of bananas)
4. `drai-hmitrötr` — Sunday (holy day)
5. `gaa-hmitrötr` — sanctuary (holy place)
6. `uma-hmitrötr` — church (holy house)
7. `ngöne-uma` — w... | Applied three random corruptions: (1) Swapped translation mappings in the Drehu↔Cemuhî table: exchanged targets so that nyine→wöli and thin→bé (swapped from original), and swapped gaa→a with ngöne-gejë→ba-jié to gaa→ba-jié and ngöne-gejë→a. Also reordered the table rows. (2) Shuffled the order of items in both Drehu an... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2008 | 4 | ### Problem 4 (20 points)
The following are words in Copainalá Zoque and their English translations:
| Copainalá Zoque | English |
|----------------------|------------------------|
| nakpatpit mis | with your cactus |
| nakpat | a cactus |
| mokpittih ... | ### Problem 4: Noun Suffixes
#### (a) Suffixes and Forms:
1. **Noun Suffixes:**
- `-k2sm2` — above, overhead.
- `-kotoya` — for.
- `-pit` — with.
2. **Morphological Changes:**
- After nasal consonants (m, n, N), stops p, t, k become voiced (b, d, g respectively).
- If k comes after y, the two sounds exc... | Applied multiple random corruptions: (1) Swapped English translations between three pairs: pokskuy <-> peroltih; k2m2NdaPm <-> koc2ktaPm; pahsungotoya <-> kumgukyotoyataPm. (2) Random English substitutions: changed 'only with the corn' to 'only with the moon' for mokpittih; changed 'right above the post' to 'right abov... | [
"swap_translations",
"random_english_substitution",
"swap_source_words"
] | |
2009 | 1 | ### Problem #1 (20 points)
The following are phrases in English and their translations into the Sulka language:
3 bicycles a moulang hori orom a tgiang
1 betel nut a tu a tgiang
10 yams o sngu a lo ktiëk
2 coconuts a lo vhoi a lomin
18 breadfruits o ngaitegaap a korlolo... | ## Problem 1
Here are the words from which the Sulka language constructs its numerals:
- tgiang 1, lomin 2, korlotge 3, korlolo 4, ktiëk 5, mhelom 20;
- hori orom addition, lo doubling;
- a singular, o plural (from 3 on).
Nouns have different forms for the two numbers (tu, sngu; vhoi, vuo). There are separate words... | Applied three corruptions: (1) Swapped Sulka translations for specific English pairs: '1 betel nut' ↔ '1 yam'; '2 betel nuts' ↔ '2 coconuts'; '10 breadfruits' ↔ '10 coconuts'. (2) Shuffled the order of the example lines randomly. (3) Replaced two English nouns with unrelated words: changed '3 breadfruits' to '3 bicycle... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2008 | 2 | ### Problem 2 (20 points)
The following are four excerpts from Old Norse poems composed around 900 C.E. All of them are written using the meter named dróttkvætt (lit. ‘court meter’):
**I**
1. ók at ísarnleiki
2. Jarðar sunr, en dunði . . .
**II**
1. þekkiligr með ása
2. þrymseilar hval deila.
3. bragðvíss at þat lagð... | ### Problem 2: Old Norse Poetic Structure
#### (a) Rules:
1. **Number of Syllables:** Each line contains 6 syllables.
2. **Alliteration:** As stated in the problem.
3. **Internal Rhyme:**
- Let V1, V2, ..., V6 represent the vowels in each line.
- At least one consonant immediately following V5 must immediately f... | Applied several randomized corruptions to the in-context examples while preserving the overall structure and questions. Changes:
- Excerpt II: swapped the nouns ‘þegnum’ and ‘ása’ between lines II:1 and II:5; also swapped the order of lines II:3 and II:4.
- Excerpt III: swapped the words ‘gnýstœrandi’ and ‘rausnarsamr’... | [
"swap_source_words",
"swap_random_lines",
"shuffle_order"
] | |
2008 | 5 | ### Problem 5 (20 points)
The following are sentences in Inuktitut and their English translations:
| Inuktitut | English |
|-------------------------------------------|--------------------------------------|
| 1. Qingmiq iputujait. | You ... | ### Problem 5: Inuktitut Sentence Structure
#### (a) Sentence Patterns:
1. **General Structure:**
- `X-(q) V–` — `X V (himself).`
- `X-(q) Y-(r)mik V-si–` — `X V a Y.`
- `X-up Y-(q) V–` — `X V the Y.`
2. **Noun Endings:**
- `-q` for definite object or subject (no definite object).
- `-r-mik` for indefin... | Applied three corruptions: (1) Swapped English translations for three random pairs of original lines: 2↔9, 4↔10, and 7↔11. (2) Shuffled the order of the 12 example rows into a new random sequence while keeping numbering 1–12. (3) Randomly substituted two English words with unrelated ones: 'dog'→'fish' in the translatio... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2009 | 2 | Problem #2 (20 points). The following are words of the Maninka and Bamana languages written in the N'Ko and the Roman script and their English translations:
(a)
| N'Ko | Roman | English |
|-------------------|----------------|---------------------------------... | ## Problem 2
The N'Ko script is written and read from right to left. The script is an alphabet: each letter stands for a consonant or a vowel. The letters within a word are joined.
(a) A tilde above a vowel letter means low tone, its absence means high tone. But a vowel has middle tone if it is marked in the same way... | - Swapped English glosses between two pairs in each table: (a) bálákàwúlì ↔ kúmáyìrà; (b) márajàba ↔ bàsítéme.
- Swapped source word pairs (N'Ko + Roman) between two rows: (a) jàtúrú ↔ kòlìjé; (b) jílàsama ↔ kɔ̀rɔkarasí (with English staying in place).
- Swapped the positions of random lines: (a) moved tùbabumóri to ro... | [
"swap_translations",
"swap_source_words",
"swap_random_lines",
"random_english_substitution"
] | |
2009 | 4 | ```markdown
Problem #4 (20 points). Given are Old Indic word stems which are thought to preserve the most ancient (Indo-European) position of the stress. They are divided into root and suffix by a hyphen. The stressed vowel bears the mark ˊ.
| stó-ma- | desire | dhūmá- | wolf | pī-tí- | drinking... | ## Problem 4
| If the stop consonant in the root | and the vowel in the suffix is a | and the vowel in the suffix is i, |
|-----------------------------------|----------------------------------|----------------------------------|
| is voiced | the stress is on the suffix. | the stress is on... | - Shuffled the order of the example stems in the initial table while keeping the same 3-pair-per-row layout.
- Swapped several glosses between stems: vŕk-a- ↔ dhūmá-; stó-ma- ↔ kāmá-; dṛ́-ti- ↔ vadh-á-.
- Replaced two English glosses with unrelated words: bhŕ̥mi- ‘mobile’ → ‘cucumber’; gā-tí- ‘walk’ → ‘giraffe’.
- Left... | [
"shuffle_order",
"swap_translations",
"random_english_substitution"
] | |
2009 | 5 | ```markdown
Problem #5 (20 points). The following are sentences in Nahuatl and their English transla-
tions:
1. nimitztlazohtla you shout
2. tikmaka in āmox̱tli you give the fish to him
3. nitalhtoā the merchant prepares something
4. kāltiā in pochtekatl in kuahwāxīnki the merchant makes the carpenter drink;
the ca... | ## Problem 5
The Nahuatl sentences begin with the predicate. The subject and object (or objects) follow in any order, preceded by *in* (a definite article).
The verb receives the following prefixes:
- subject: *ni-* 1st person sg, *ti-* 2nd person sg, —— 3rd person sg;
- object: *nēch-* 1st person sg, *mitz-* 2nd pe... | Applied three random corruptions: (1) Swapped English translations between lines 1↔5 and 3↔16. (2) Swapped positions of source-language nouns/words within lines 4 (kuahwāxīnki↔pochtekatl), 10 (kuahwāxīnki↔āmox̱tli), and reordered words in 16 (brought 'in pochtekatl' before the verb). (3) Randomly replaced two English n... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2009 | 3 | ### Problem #3 (20 points)
Here are the names of 24 Burmese children and their dates of birth:
| boys | date of birth | girls | date of birth |
|------------------|---------------|---------------|---------------|
| cɔ wiɴ auɴ | 07/08/2009 | yiɲyiɲ myiɴ | 07/15/2009 |
| zeiɲa cɔ ... | ## Problem 3
We can see that the names of the children born on the same day of the week have similar first sounds:
- Monday: kauɴ myaʔ, khiɴ le nwe, khaiɴ miɴ thuɴ, keþi thuɴ
- Tuesday: zeiya cɔ, su myaʔ so, susu wiɴ, shaɴ thuɴ, shu maɴ cɔ
- Wednesday: wiɴ i muɴ, lwiɴ koko, wiɴ cɔ auɴ, yadana u, yiɴyiɴ myiɴ
- Thursda... | Applied multiple subtle corruptions: (1) Shuffled the order of several table rows. (2) Swapped some dates between specific existing entries to break learnable mappings (boys: ne liɴ↔lwiɴ koko; myo khiɴ wiɴ↔θeʔ auɴ; girls: khiɴ le nwe↔wiɴ i muɴ; mimi khiɴ↔yadana u). (3) Swapped source words within two names: 'wiɴ cɔ auɴ... | [
"shuffle_order",
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2010 | 1 | ### Problem 1 (20 points)
Given are verbs of the Budukh language in three forms:
| Form 1: prohibitive mood | Form 2: future tense | Form 3: future tense, | Translation |
| class I (masculine) | class II (feminine) | class III (neuter) | |
|---------------------------|-------------... | ### Problem 1: Morphological Rules
#### Rules:
1. **Form 1:** `-mV-` after the first vowel, where:
- `V` depends on the vowel in the following syllable:
- `a` before `a`
- `o` before `o` or `u`
- `e` before `i`
- `ö` before `ü`
2. **Form 2:**
- `-a` if the stem ends in `-aR` or `-oR`
- `-R... | Applied three random corruptions: (1) Shuffled the order of the example rows. (2) Swapped English translations between two pairs: swapped translations of the rows with amarxar…arxara (sleep) and ?…quroo1ura (bring to a halt); swapped translations of the rows with h̨ümoč. onxu…h̨ürč. onxuna (overtake) and ?…jölküla (mak... | [
"shuffle_order",
"swap_translations",
"swap_source_words"
] | |
2010 | 2 | ### Problem 2 (20 points)
Given are Drehu numerals in alphabetical order and their values in ascending order:
| Drehu | Value |
|----------------------------------|-------|
| lueatr nge luako | 78 |
| caatr nge caako | 36 |
| ekengömen nge ekaatr ... | ### Problem 2: Numeral System
#### Rules:
1. **Numbers 1–4:** `caa 1, lue 2, köni 3, eke 4`
2. **Multiples of 5, 10, 15:** `β-pi = 5β` (1 ≤ β ≤ 3)
3. **Numbers 6–9, 11–14, 16–19:** `α-ngömen = 5 + α`, `α-ko = 10 + α`, `α-qaihano = 15 + α` (1 ≤ α ≤ 4)
4. **Multiples of 20, 40, 60, 80:** `γ-atr = 20γ` (1 ≤ γ); e.g., `caa... | - Swapped translation pairs (values) for two entries: (1) 'caatr nge caako' now 36 and 'caatr nge caaqaihano' now 26; (2) 'köniatr nge köniko' now 89 and 'lueatr nge luepi' now 50.
- Shuffled the order of example lines so the table no longer matches the stated alphabetical/ascending pairing.
- Swapped source word order... | [
"swap_translations",
"shuffle_order",
"swap_source_words"
] | |
2010 | 4 | ### Problem 4 (20 points)
One of the major achievements in genetics was the decipherment of the genetic code—the creation of an mRNA–polypeptide calendar. Polypeptides (proteins) are building kites of all living organisms. Polypeptide molecules are chains that consist of amino acids (denoted as Arg, Leu, Phe etc.), and... | ### Problem 4: mRNA and Amino Acid Sequence
#### (a) mRNA Sequence:
1. `AUG` — Met
2. `UUA` — Leu
3. `ACG` — Thr
4. `UUC` — Phe
5. `UAA` — stop
1. `AUG` — Met
2. `UGG` — Trp
3. `GGG` — Gly
4. `GGA` — Gly
5. `CAC` — His
6. `CAG` — Gln
#### (b) Possible Combinations:
1. Met-Lys-Cys-Ile ← `AUG`
- `AAA`/`AAG`
- `UG... | Applied three corruptions to the in-context examples while leaving the assignment questions (a), (b), (c) unchanged. 1) Swapped amino acid labels in the example polypeptides only: Arg↔Gly, Leu↔Val, Lys↔Glu. 2) Shuffled the order of the four example polypeptide lines (new order: original 3rd, 1st, 4th, 2nd). 3) Randomly... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2010 | 5 | ### Problem 5 (20 points)
Given are words of two dialects of the Romansh language and their English translations. Some cells have been left blank:
| Sursilvan | Engadine | English |
|---------------|--------------|----------------------|
| tut | tuot | all |
| ur... | ### Problem 5: Dialectal Variations in Vowel Usage
#### (a) Translations:
1. `uo` before a cluster of `l` or `r` and another consonant.
2. `u` before `l` or `r` without another consonant.
3. `u` → `o` before `m`.
4. `u` → `uo` before another consonant.
| **Sursilvan** | **Engadine** |
| | Applied three random corruptions: (1) Swapped English glosses between buglia/buoglia and discuors/discuors, and between angul/angul and fuorma/fuorma. (2) Swapped Sursilvan forms between the rows for 'name' and 'world' (num ↔ mund), and swapped Engadine forms between the rows for 'tooth filling' and 'to show' (plomba ↔... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2011 | 1 | **Problem #1 (20 points).** Given are verb forms of the Menominee language as well as their English translations:
| Menominee | English Translation |
|-----------------|----------------------------------------------------------|
| kewāpənæhkæq | we_1+2 begin to fish ... | ## Problem 1
The Menominee verb forms have the following structure:
| | kaw | down |
|-------|------|-------|
| | ket | out |
| | kēsk | through |
| | pahk | off |
| | pāhk | open |
| | pīt | hither |
| | taw | pierce |
| | wack | around |
| | wǣp | beg... | Applied three corruptions: (1) Swapped English translations between specific Menominee forms: kewāpeqtəq ↔ wæpōhnæw, kawāham ↔ pāhkəqtaw, and ketænəm ↔ kekætəhnæq. (2) Random English substitutions: changed 'we_1+2 begin to dig' to 'we_1+2 begin to fish' for kewāpənæhkæq, and 'he walks roundabout...' to 'he sleeps round... | [
"swap_translations",
"random_english_substitution",
"shuffle_order"
] | |
2011 | 2 | Problem #2 (20 points). The following are words of the Faroese language written in the regular orthography and in phonetic transcription as well as their English translations:
| bóga | ? | hen bird |
| deyði | [deiːji] | (I) kill |
| eyður | ... | ## Problem 2
(a)
| Word | Phonetic Transcription | Translation |
|--------|------------------------|--------------------------------------|
| bøga | [bøva] | hen bird |
| knoðar | [knOar] | (he) kneads |
|... | Applied three randomized corruptions: (1) Swapped English translations between three pairs: eyður↔logi (wealth/flame), søga↔vágur (story/gulf), tregr↔vegur ((he) injures/(he) raises). (2) Swapped the Faroese source words (first column only) in three pairs while keeping the original transcriptions/translations in place:... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2011 | 3 | ```markdown
Problem #3 (20 points). Given are phrases in the Vai language as well as their English translations:
| Vai Phrase | English Translation |
|-----------------------------|--------------------------------|
| mùsù jáŋ́ɛ́ lɔ̀-kàì | the tall woman’s brother |
| kaíɛ́ á l... | ## Problem 3
Rules:
1. Adjectives follow their nouns.
2. A noun (or the adjective if there is one) gets the marker -Ě, unless it is inalienably possessed (body part, kinship term); in the latter case it is preceded by the possessor.
3. Alienable possession is expressed by á between the possessor and the possessed.
4... | Applied three corruptions: (1) Shuffled the order of the nine example rows. (2) Swapped English translations between three pairs of Vai phrases: kaíɛ́ á lɛ̀ndɛ́ɛ́ ↔ gbòmùé á nyìmiɛ̀; nyìmiì jáŋ́ á gbòmù-lɛ̀ndɛ̀ ↔ kɔ̀àŋ̀jà lɔ̀ɔ̀ kɛŋ̀ì; kaíɛ́ káfà ↔ kàndɔ́ jáŋ́ɛ́. (3) Replaced two English words with unrelated ones: 'fa... | [
"shuffle_order",
"swap_translations",
"random_english_substitution"
] | |
2011 | 4 | ### Problem #4 (20 points)
Given are words in Nahuatl as well as their English translations in arbitrary order:
**coltzintli**, **conehuah**, **achilli**, **totoltetl**, **capilconehuah**, **oquichhuah**, **atl**, **oquichconetl**, **chilli**, **callah**, **tēhuah**, **acalhuah**, **conetl**, **totoloquichtzintli**, ... | ## Problem 4
(a) Determine the correct correspondences.
| Nahuatl | Translation |
|-------------------|--------------------------------|
| a-cal-huah | canoe owner (a-cal-li canoe, “water house”) |
| a-chil-li | water pepper |
| a-tl | water ... | Applied multiple corruptions: (1) Shuffled the order of the Nahuatl word list and independently shuffled the English gloss list. (2) Swapped internal morpheme order in two Nahuatl compounds to disrupt learnable patterns: 'conehuahcapil' -> 'capilconehuah' and 'oquichtotoltzintli' -> 'totoloquichtzintli'. (3) Random Eng... | [
"shuffle_order",
"swap_source_words",
"random_english_substitution"
] | |
2012 | 1 | ```markdown
**Problem #1 (20 points).** Here are some sentences in the central dialect of Dyirbal as well as their English translations:
1. bayi yaɽa ŋuɳaymuɳa baŋgu guruguŋu biŋgunman.
Booze is making the man that is always being blamed tired.
2. balan yabu bimabaŋaɲaɽaymuɳa baŋgul vaɽaŋgu guliŋgu yuɲaɲu.
... | ## Problem 1
**Word order**: OSV (O: object, S: subject, V: verb), NA (N: noun, A: adjective).
**A → V** ('make A'): A-man.
**V → A transformation**:
| V | 'that is always being Ved' | 'that is always Ving Ns' |
|-----|----------------------------|--------------------------|
| -n | -l-muŋa | N... | Applied multiple random corruptions to the in-context examples (1–16) while leaving the questions (a–c) unchanged. Swapped English translations between pairs of examples: (3↔5), (8↔12), and (10↔14). Swapped two Dyirbal source words within three sentences: in 4 swapped 'yila' and 'wura'; in 7 swapped 'midin' and 'bimaŋg... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2012 | 3 | ```markdown
**Problem #3 (20 points).** Here are some sentences in Basque as well as their English translations in arbitrary order. One of the English sentences corresponds to two sentences in Basque:
mintzatu gatzaizkizue, hurbildu zaizu, lagundu naute, lagundu ditugu, ahaztu ditut, ukitu nauzue, lagundu zaizkigu, ah... | ## Problem 3
**Verb Agreement Table:**
| 1st person sg | 1st person pl | 2nd person sg | 2nd person pl | 3rd person sg | 3rd person pl |
|---------------|---------------|---------------|---------------|---------------|---------------|
| nau- | | | | du- | d... | Applied three random corruptions: (1) shuffled the order of both the Basque and English example lists; (2) swapped source-language verbs between a few Basque examples (ahaztu/hurbildu and lagundu/ukitu swaps, plus lagundu/mintzatu) to break consistent mappings while keeping forms plausible; (3) replaced two English ver... | [
"shuffle_order",
"swap_source_words",
"random_english_substitution"
] | |
2012 | 2 | ```markdown
Problem #2 (20 points).
| Umbu-Ungu | Umbu-Ungu |
|----------------------|------------------------------------|
| 30 polangipunga talu| 79 tokapu yepoko polangipunga talu|
| 27 tokapunga telu | 69 tokapu talu tokapunga telu |
| 20 supu | 48 tokapu... | ## Problem 2
**Number Word Table:**
| Number | Umbu-Ungu |
|--------|-----------|
| 1 | telu |
| 2 | talu |
| 3 | yepoko |
| 12 | rurepo |
| 16 | malapu |
| 20 | supu |
| 24 | tokapu |
| 28 | alapu |
| 32 | polangipu |
**Mathematical Formulas:**
... | Applied three independent corruptions: (1) swapped several translation pairs: 21↔27 (left column phrases), 40↔50 and 48↔97 (right column phrases); (2) swapped source-language word order inside a few examples: 'rureponga talu'→'talu rureponga', 'malapunga yepoko'→'yepoko malapunga', and 'tokapu talu polangipunga yepoko'... | [
"swap_translations",
"swap_source_words",
"shuffle_order"
] | |
2012 | 4 | ```markdown
Problem #4 (20 points). One linguist decided to write a grammar of the Teop language. First she asked her informants to translate separate sentences into their mother tongue. Here is what she got:
1. You (sg.) struck me.
*Ean paa tasu anaa.*
2. He ate the fish.
*Enaa paa tara a taba’ani.*
3. We... | ## Problem 4
The sentences have the following structure: \( S \ \text{paa} \ V \ O \ [O'] \) (S: subject, V: verb, O: object, \( O' \): another object).
| | give | call | strike | kill |
|--------|------|-------|--------|-------|
| O | to whom | whom | whom | whom |
| \( O' \) | what | what | with ... | - Swapped English-Teop pairings between sentences 2↔6 and 4↔7 to misalign translations.
- Swapped two Teop content words within sentences 8 (otei ↔ overe) and 12 (visoasi ↔ ororara).
- Randomly substituted English words: changed “child” to “window” in sentence 3, and “axe” to “umbrella” in sentence 11.
- Exchanged the ... | [
"swap_translations",
"swap_source_words",
"random_english_substitution"
] | |
2013 | 1 | Problem #1 (20 points). Here are some words in Yidiny and their English translations:
bamaɡimbal without a woman biŋiɲiːnmu from a hornet
buŋa woman ɡaɟaɡimbaːm from a white man
mulariːn of an initiated man ɟuduːlumuɟaɲ with a scissors
muyubara ... | ## Problem 1
1. Rules:
- If the number of syllables in the word (= stem + ending) is even, all syllables are short. If the number of syllables in the word is odd, the last even-numbered syllable of the stem is lengthened.
| Syllables | Example |
|-----------|--------------------------|
| ... | Applied three random corruptions: (1) swapped translations between three pairs: gudaːga↔muyubara (dog↔stranger), bimbiːn↔muɲam (of a father↔mother), and bamaɡimbal↔buɲaɡimbal (without a person↔without a woman); (2) shuffled the order of the example lines; (3) replaced two English glosses with unrelated words: tortoise→... | [
"swap_translations",
"shuffle_order",
"random_english_substitution"
] | |
2012 | 5 | **Problem #5 (20 points).** Here are some words and word combinations in Rotuman as well as their English translations:
| Rotuman | English | Rotuman | English |
|------------|-------------------------|--------------|----------------------------|
| lala | shallow ... | ## Problem 5
When two words constitute a phrase, the form of the first word undergoes the following change:
-VCV → -VC
-VCu → -VC
-VCi → -VC
-VCa → -VaC
(V: vowel, C: consonant).
The same thing happens when an adjective is formed by doubling a noun or a verb: ‘ele + ‘ele ‘to be near × 2 = shallow’.
The wor... | Applied multiple subtle corruptions: (1) Swapped English glosses between several Rotuman items: ‘olo↔niu (to cut/copra), huag ‘el‘ele↔huag to‘a (impatient/courageous), pogi↔pala (night/pierce). (2) Random English substitutions: changed hün kia’s meaning to 'sidewalk' and kȧlu’s meanings to 'umbrella; to levitate'. (3)... | [
"swap_translations",
"random_english_substitution",
"swap_source_words",
"swap_random_lines"
] | |
2013 | 2 | Problem #2 (20 points). Here are some words in Tundra Yukaghir and their English translations in arbitrary order:
```
joqodile, aarriŋjohul, ileninme, saadoŋoʒ, joqol, møŋer, johul,
wødu, joqonchoje, aariinmøŋer, cireme, joqoninme, onʒoʒ,
wodawur, ilenlegul, johudawur, ciremennime, saanchojo
```
wooden knife, cradl... | ## Problem 2
(a)
| Tundra Yukaghir | English |
|-----------------|-----------------|
| ilennime | herd of deer ("house of deer") |
| joqonnime | wooden house ("Yakut house") |
| saancohoje | wooden knife |
| johudawur | nose case |
| ilenlegul | deer feed |
| cirem... | Applied two corruptions to the training data while keeping the problem structure and questions intact: (1) randomly shuffled the order of the Yukaghir word list and the English gloss list; (2) replaced two English glosses with unrelated nouns: 'grandchild' → 'helicopter' and 'nest' → 'pumpkin'. | [
"shuffle_order",
"random_english_substitution"
] | |
2013 | 4 | ### Problem #4 (20 points)
Here are some sentences in Muna and their English translations:
1. **murihino andoandoke dofoni we molo.**
The ants’ houses are warm.
2. **lambuku nakumodoho.**
My guitar will be far.
3. **lambuhindo lagahi nofanaka.**
The Monkey’s pupils are climbing the mountain.
4. **la... | ## Problem 4
The word order is \[ subject \, verb \, object \]. If the subject is neither possessed nor a proper name, it is preceded by the article *a*.
Noun: \[ \text{root} + \left[ \text{-hi plural} \right] +
\begin{array}{|c|}
\hline
\text{possessor:} \\
-ku \, \text{1st person sg} \\
-no \, \text{3rd person sg}... | Applied multiple subtle corruptions to the learning examples (1–11) while leaving the assignment questions (12–17) unchanged. Specifically: (1) Swapped English translations between lines 1↔3 and 5↔8. (2) Swapped the positions of example lines 4 and 10 (moving both Muna and their paired English lines). (3) Made random E... | [
"swap_translations",
"swap_random_lines",
"random_english_substitution",
"swap_source_words"
] | |
2013 | 3 | ### Problem #3 (20 points)
Here are some words and phrases in Pirahã and simplified transcriptions of their pronunciation in normal speech:
*bagiai baabi* ba.gja.¹baa.bi bad thief
*kaoaibogi* kao.¹ai.bo.gi jungle spirit
*bahoigatoi baihiigi* ba.₂hoi.ga.to.bai.¹hii.gi ... | ## Problem 3
Rules:
1. Ɂ [ʔ].
2. A noun and a following modifier are pronounced as one word, but at the end of the first word of the phrase [i] after a vowel is lost and in the beginning of the second word [ʔ] is lost.
xisitai xagai [ʔisitaɁí_ʔxagai] → [ʔisitaagái]
3. 1 syllable = CVV, CV or VV (C = consonant, ... | Applied multiple random corruptions: (1) Swapped English meanings of several items: giopai↔hixi (dog/rat), xogiai↔xiga (big/hard), kagihi↔xabagi (wasp/toucan). (2) Random English substitutions: changed tagasaga from 'machete' to 'turtle' and xiiaapisi from 'sleeve' to 'cloud'. (3) Swapped positions of several example l... | [
"swap_translations",
"random_english_substitution",
"swap_random_lines"
] | |
2014 | 1 | ### Problem #1 (20 points)
Here are some verb forms of Benabena and their English translations:
*nofine* you_sg are piercing him
*lahala'ibe* you both will strike us
*nokoho'ibe* we will strike you_sg
*lenahalube* I will dance you_pl
*nohobe* I am striking him
*lenifilu'ibe*... | ## Problem 1
Rules:
1. present: **no-**
2. object =
| | singular | plural |
|--------|----------|--------|
| 1 | nV- | lV-. |
| 2 | kV- | lenV-. |
V is the vowel from the following syllable
3.
| | ‘strike’ | ‘pierce’ |
|----------------|---------... | Applied three randomized corruptions: (1) Shuffled the order of the 15 example lines. (2) Swapped English translations between three pairs: 5↔8, 2↔3, and 6↔13. (3) Randomly substituted English verbs in two translations: changed 'I will strike you_pl' to 'I will dance you_pl' (line 10) and 'because we both are piercing ... | [
"shuffle_order",
"swap_translations",
"random_english_substitution"
] | |
2013 | 5 | ### Problem #5 (20 points)
In a series of experiments run in Carnegie Mellon University (Pittsburgh, USA) in 2010, volunteers were first shown some English words, while activity was being registered in different locations of their brains. Then the volunteers were asked to think of some other words from a preselected l... | ## Problem 5
Location A is activated by the idea of shelter. Location B is activated by the idea of manipulation. Location C is activated by the idea of eating. Location D is activated by long words. The researchers claim that the first three factors have high ecological validity (i.e., the results of the experiment c... | Applied multiple small corruptions to the first (training) table only: (1) Swapped source word labels between 'arm' and 'cup', and between 'igloo' and 'lettuce' so their activity patterns are reassigned. (2) Swapped translations for 'airplane'↔'corn' and 'key'↔'apartment'. (3) Randomly substituted translations: set 'sc... | [
"swap_source_words",
"swap_translations",
"random_english_substitution",
"shuffle_order"
] | |
2014 | 2 | Problem #2 (20 points). Here are the singular, dual and plural forms of some Kiowa nouns and their English translations. Not all forms are given, although all exist.
| singular | dual | plural | |
|--------------|---------------|----------------|---------------------|
| aɗɔ̀ ... | ## Problem 2
### Rules:
| | singular | dual | plural |
|---------------|----------|------|--------|
| people, animals, tools | -∅ | -n | -gɔ |
| fruit | -gɔ | -∅ | -gɔ |
| all other things | -dɔ | -n | -dɔ |
### Answers:
| singular | dual | plural | Trans... | Applied three corruption types: (1) Swapped English gloss pairs: tree↔spoon, knife↔hat, orange↔lemon. (2) Random English substitutions: replaced 'bread' with 'mountain' and 'chair' with 'airplane'. (3) Swapped source forms: exchanged the plural forms of the 'knife' and 'hat' rows (kʼɔ̀gɔ̃ ↔ kʼɔnbohon) and swapped the d... | [
"swap_translations",
"random_english_substitution",
"swap_source_words"
] | |
2014 | 5 | Problem #5 (20 points). Here are some words and word combinations in Northwest Gbaya and their English translations in arbitrary order:
ʔáá, ʔáá náŋ nú kò, ʔáá lébé, búmá yíʔ, búmá wí yíʔ, dáŋ zòk, dí fô, dí sèè, kò yíʔ, kò gòk, náŋ zù, nú fô, nú sèè, sèè wí, yíʔ, yíʔ zù, zù
poisonous snake; at the surface; bicycle; ... | ## Problem 5
(a)
| Northwest Gbaya | Translation |
|---------------------|------------------------------------|
| ʔáá | to put |
| ʔáá náng nú kò | to be dying |
| ʔáá sèè | to envy ... | Applied three random corruptions: (1) swapped/replaced source morphemes inside several Northwest Gbaya items (e.g., búmá zù yíʔ → búmá wí yíʔ; náŋ wí → náŋ zù; yíʔ wí → yíʔ zù; dáŋ gòk → dáŋ zòk; kò zòk → kò gòk; ʔáá sèè → ʔáá lébé; nú lébé → nú sèè), (2) randomly substituted two English glosses with unrelated words (l... | [
"swap_source_words",
"random_english_substitution",
"shuffle_order"
] | |
2014 | 3 | ### Problem #3 (20 points)
Once upon a time in the Tangut Empire ("the Great State of the White and the Lofty") there lived two brothers and two sisters. They each had a piano and a daughter.
Family tree (△ = man, ○ = woman):
△───△───○───○ ← 2 brothers and 2 sisters (siblings)
│ │ │ │
... | ## Problem 3
### (a)
```
Lhie²nyn²
/ \
Wa²nie¹ Syn¹mei¹
/ \
Ke.i¹¹ey² Sei¹na¹
/ \
Sie¹tsie¹ Ldiu²śe¹
/ \
Œan¹nia¹ Nie²tse.¹
```
### (b)
Nie²tse.¹ Sie¹tsie¹ 'yn¹ ndo.n¹ ngu²... | Applied three corruption strategies: (1) Swapped source words (names) within several sentences: line 2 (Lhie²nyn² ↔ Ngwi¹mbyn²), line 14 (Lhie²nyn² ↔ Kęi¹źey²), line 28 (Śan¹nia¹ ↔ Sie¹tsie¹), line 31 (Śan¹nia¹ ↔ Nie²tsę¹). (2) Swapped random lines: exchanged contents of lines 10↔19, 5↔22, and 23↔33. (3) Random English... | [
"swap_source_words",
"swap_random_lines",
"random_english_substitution"
] |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 7