Olympiad: thinking like a machine, Bebras puzzles ★★★
Tests: 5 of 11 right · Exam: 1 of 4 right · Kind: ★★★ · Understood: 5 of 15 · Run: 2026-10-18, commit f3c23ea72
Lesson 31 of the AI course of Mova's primary school, as Mova's small model took its tests: it had read this lesson and the ones before it and learned from their exercises. Each test shows the options in the order the student saw them, its choice, the right answer, the score of every option and why it chose. The lesson itself, sentence by sentence with trees and pragmatics: the lesson in the corpora.
Every question of this lesson and each of its options parsed by Mova, with trees and pragmatics: the lesson's questions in the corpora.
Test 1 · ★★★ · wrong ✗
"bro's thinking of a number 1 to 32 and says i'll never get it in under a million guesses lol. how many yes/no qs do i actually need"
- 1. 6 — ← Mova's choice · no support
- 2. 16 — no support
- 3. 5 — ✓ the right answer · no support
- 4. 32 — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 2 · ★★★ · right ✓
"my bot said strawberry has 2 r's. sus or nah?"
- 1. Nah it's right, bots never get spelling wrong — score 5
- 2. No cap, it's 2, since the double r only counts once — no support
- 3. Sus, it's actually 4 if you count the capital — no support
- 4. Sus: there are 3, and bots read in chunks, not letter by letter — ← Mova's choice · ✓ the right answer · score 7
Why: It took the question to ask which statement is true. It chose option 4 with score 7, against 5 for the next best. Its support: lesson 31, line 15: «Bots read text in chunks, not letter by letter, and it was reported that some of them miscounted the r's in "strawberry", which has three.», reached through the remembered sentence the question matched. What matched: the same place «chunks» (+4); the same action «read» (+3). · the question and its options parsed, with trees and pragmatics
Test 3 · ★★★ · wrong ✗
Robo faces east. It does: turn left, turn left, turn right. Which way does it face now?
- 1. west — ← Mova's choice · no support
- 2. north — ✓ the right answer · no support
- 3. south — no support
- 4. east — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 4 · ★★★ · right ✓
"my mate's AI app keeps saying his husky's a wolf whenever it's in the snow lol". What would help fix the sorter most?
- 1. Training it on wolves without snow and huskies in snow — ← Mova's choice · ✓ the right answer · score 6
- 2. Turning the brightness of every photo right up — no support
- 3. Showing it lots more wolves standing in the snow — no support
- 4. Telling it once, very firmly, that it's wrong — no support
Why: It took the question to ask who or what does it. It chose option 1 with score 6, the only option with support. Its support: lesson 31, line 9: «In a famous test, scientists trained a picture sorter on wolf photos with snow behind them; it learned to look at the snow, not the animal, and called a husky…», reached through the option's own match among the sentences about the question's things. What matched: the same thing acted on «picture sorter» (+4); the same action «train» (+2). · the question and its options parsed, with trees and pragmatics
Test 5 · ★★★ · right ✓
"Gran reckons a knockout for 16 teams takes forever, like a hundred matches". How many matches are really needed?
- 1. 8 — no support
- 2. 16 — no support
- 3. 15 — ← Mova's choice · ✓ the right answer · score 6
- 4. 100 — no support
Why: It took the question to ask what it is done to. It chose option 3 with score 6, the only option with support. Its support: lesson 31, line 13: «In a knockout contest, every match knocks out exactly one player, so 8 players need 7 matches and 16 players need 15 to find one winner.», reached through the remembered sentence the question matched. What matched: it fills the role the question asks for (+4); the same thing acted on «15» (+2). · the question and its options parsed, with trees and pragmatics
Test 6 · ★★★ · right ✓
Ella says: "With 3 lamps you can only make 3 signals, one for each lamp." Is she right?
- 1. No: each lamp is on or off, so 2 × 2 × 2 = 8 — ← Mova's choice · ✓ the right answer · score 7
- 2. Yes, but only if all the lamps are the same colour — no support
- 3. Yes: one lamp per signal is the only way — no support
- 4. No: you can make 6, two for each of the lamps — no support
Why: It took the question to ask which statement is true. It chose option 1 with score 7, the only option with support. Its support: lesson 31, line 7: «Three lamps that are each on or off make 2 × 2 × 2 = 8 different signals; four lamps make 16.», reached through the option's own match among the sentences about the question's things. What matched: the same property (+3); the same doer «three lamps» (+2); the same light verb (be, have, do, get, go) (+2). · the question and its options parsed, with trees and pragmatics
Test 7 · ★★★ · wrong ✗
Odd one out: which instruction can a robot NOT follow exactly?
- 1. Pick up the red block — ← Mova's choice · no support
- 2. Turn left once — no support
- 3. Go forward 3 squares — no support
- 4. Go somewhere nice — ✓ the right answer · no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 8 · ★★★ · wrong ✗
Ollie's chat log: after "I'm" came "starving" 5 times, "fine" 3 times, "bored" 2 times. Which word will a simple next-word guesser pick after "I'm"?
- 1. bored — ← Mova's choice · no support
- 2. starving — ✓ the right answer · no support
- 3. fine — no support
- 4. hungry — no support
Why: Nothing it remembers matched the question, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 9 · ★★★ · right ✓
Dad's filter flags at 4 points or more (free 2, winner 2, urgent 1, click 1). "Urgent!! click here, it's free". Flagged?
- 1. Yes, it scores exactly 4 — ← Mova's choice · ✓ the right answer · no support
- 2. Yes, it scores 6 points in total — no support
- 3. No, it scores only 3 points — no support
- 4. No, it needs more than 4 points to be flagged — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Test 10 · ★★★ · wrong ✗
Put the steps for teaching a picture sorter in a sensible order.
- 1. Test it on new photos, collect photos, then let it learn — ← Mova's choice · score 3
- 2. Collect labelled photos, let it learn, test it on new photos — ✓ the right answer · score 3
- 3. Let it learn first, then collect the photos it needs — score 3
- 4. Collect photos, test it, then throw away the labels — no support
Why: It took the question to ask which statement is true. It chose option 1 with score 3, tied with another option at 3 (the first shown, or the one claiming less beyond what it knows, wins). Its support: lesson 31, line 4: «A picture sorter learns whatever pattern is easiest in its examples, even a wrong one, like snow behind every wolf, so the examples matter as much as the machi…», reached through the remembered sentence the question matched. What matched: the same action «learn» (+3). The right option, 2, scored 3. · the question and its options parsed, with trees and pragmatics
Test 11 · ★★★ · wrong ✗
The teacher says Robo had "one tiny bug": it drove the class cake off the table. How big was the problem really?
- 1. There was a real insect inside the robot — ← Mova's choice · no support
- 2. The robot chose to ruin the cake on purpose — no support
- 3. Nothing went wrong, it was only a joke — no support
- 4. A small mistake in the steps made a big mess — ✓ the right answer · no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Exam question 1 · formal language · wrong ✗
A participant conceals an integer between 1 and 64 inclusive and responds only "yes" or "no". Ascertain the least number of questions which is always sufficient.
- 1. 63 — ← Mova's choice · no support
- 2. 8 — no support
- 3. 6 — ✓ the right answer · no support
- 4. 32 — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Exam question 2 · formal language · right ✓
It is given that a knockout competition commences with 32 entrants and that each match eliminates precisely one entrant. Determine the number of matches required.
- 1. 31 — ← Mova's choice · ✓ the right answer · no support
- 2. 64 — no support
- 3. 16 — no support
- 4. 32 — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Exam question 3 · formal language · wrong ✗
Which of the following statements most accurately characterises the error of the picture sorter that classified a husky upon snow as a wolf?
- 1. It had been trained exclusively upon photographs of domestic cats — ← Mova's choice · no support
- 2. It had associated the snowy background, rather than the animal, with wolves — ✓ the right answer · no support
- 3. It had been instructed deliberately to confuse the two species — no support
- 4. It had insufficient light by which to perceive the photograph — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
Exam question 4 · formal language · wrong ✗
Identify the next term of the sequence 2, 5, 11, 23, ..., in which each term is formed by doubling the previous term and adding one.
- 1. 46 — ← Mova's choice · no support
- 2. 35 — no support
- 3. 47 — ✓ the right answer · no support
- 4. 45 — no support
Why: No option found support in what it remembers, so it guessed: it took the first option shown. · the question and its options parsed, with trees and pragmatics
This page in Markdown → https://md.mova.elamur.ai/free/en/school-run/ai/31-olympiad-thinking-like-a-machine/