Dataset Viewer
Auto-converted to Parquet Duplicate
data_source
stringclasses
4 values
prompt
listlengths
2
2
ability
stringclasses
2 values
reward_model
dict
extra_info
dict
sub_source
stringclasses
17 values
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Virat is a guy who likes solving problems very much....
code
{ "ground_truth": "{\"inputs\": [\"2\\n6 1 15\\n10 17 29\"], \"outputs\": [\"11\\n9\"]}", "style": "rule" }
{ "difficulty": "UNKNOWN_DIFFICULTY", "id": "taco-cft-UNKNOWN_DIFFICULTY-1492", "rating": null, "split": "confidence_train", "sub_source": "UNKNOWN_DIFFICULTY" }
UNKNOWN_DIFFICULTY
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "You are developing software to stream music to users...
code
{ "ground_truth": "{\"inputs\": [\"3 20\\n50 70 10\\n100 110 4\\n150 190 16\\n\", \"4 100\\n0 50 100\\n0 50 100\\n0 50 100\\n1000 1500 100\\n\"], \"outputs\": [\"5\\n\", \"-17\\n\"]}", "style": "rule" }
{ "difficulty": "HARD", "id": "taco-cft-HARD-1474", "rating": null, "split": "confidence_train", "sub_source": "HARD" }
HARD
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "Given that $\\tan α$ and $\\tan β$ are the two real roots of the equation $2x^{2}+3x-7=0$, find the value of $\\tan (α+β)$ .", "role": "user" } ]
math
{ "ground_truth": "-\\frac{1}{3}", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-2464", "rating": null, "split": "confidence_train", "sub_source": "cn_k12" }
cn_k12
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Chef likes inequalities. Please help him to solve ne...
code
{ "ground_truth": "{\"inputs\": [[\"1\", \"2 3 3 4\"], [\"1\", \"2 999999 1 1000000\"]], \"outputs\": [[\"3\"], [\"499998500001\"]]}", "style": "rule" }
{ "difficulty": "UNKNOWN_DIFFICULTY", "id": "taco-cft-UNKNOWN_DIFFICULTY-504", "rating": null, "split": "confidence_train", "sub_source": "UNKNOWN_DIFFICULTY" }
UNKNOWN_DIFFICULTY
codecontests
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Summer holidays! Someone is going on trips, someone ...
code
{ "ground_truth": "{\"inputs\": [\"4 1\\n0 2\\n0 3\\n3 5\\n0 6\\n\", \"4 2\\n2 1\\n3 5\\n2 3\\n1 5\\n\", \"2 1\\n2 4\\n4 4\\n\", \"2 1\\n10 11\\n9 20\\n\", \"2 1\\n3 7\\n5 7\\n\", \"2 1\\n1 2\\n1000 1000\\n\", \"2 1\\n14 28\\n15 28\\n\", \"3 1\\n10 11\\n12 13\\n8 10\\n\", \"2 1\\n5 7\\n6 7\\n\", \"3 1\\n1 4\\n2 4\\n3...
{ "difficulty": null, "id": "810_B. Summer sell-off", "rating": 1300, "split": "confidence_train", "sub_source": "rating_1300" }
rating_1300
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "In a distant realm, all Snarks are Garbles, all Twerks are Garbles, all Snarks are Quarks, and all Quarks are Twerks. Which of the following statements is necessarily true based ...
math
{ "ground_truth": "\\textbf{(C)}", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-3109", "rating": null, "split": "confidence_train", "sub_source": "synthetic_amc" }
synthetic_amc
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Let G be a simple graph. Let W be a non-empty subset...
code
{ "ground_truth": "{\"inputs\": [\"4\\n1 2\\n2 3\\n3 4\\n\", \"18\\n1 14\\n1 10\\n1 4\\n1 2\\n1 5\\n3 9\\n4 11\\n6 10\\n6 18\\n7 9\\n8 18\\n9 14\\n10 12\\n10 13\\n12 16\\n12 17\\n15 16\\n\", \"4\\n1 2\\n1 3\\n1 4\\n\", \"17\\n1 17\\n1 9\\n1 3\\n2 6\\n3 4\\n3 12\\n5 16\\n5 11\\n6 17\\n7 14\\n7 8\\n9 13\\n10 17\\n14 17...
{ "difficulty": "VERY_HARD", "id": "taco-cft-VERY_HARD-1939", "rating": null, "split": "confidence_train", "sub_source": "VERY_HARD" }
VERY_HARD
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "In this problem, you are required to write a program...
code
{ "ground_truth": "{\"inputs\": [\"5\\n3 C E G\\n3 C D G#\\n4 C A G E\\n5 F A C E D\\n3 C D E\", \"5\\n3 C E G\\n3 C D G#\\n4 D A G E\\n5 F A C E D\\n2 C D E\", \"5\\n3 D E G\\n3 C D G#\\n4 D A G E\\n5 F A C E D\\n2 C D E\", \"5\\n3 C E G\\n3 C E G#\\n4 C A G E\\n5 F A C E D\\n3 C C E\", \"5\\n3 C E F\\n3 C D G#\\n4 ...
{ "difficulty": "UNKNOWN_DIFFICULTY", "id": "taco-cft-UNKNOWN_DIFFICULTY-755", "rating": null, "split": "confidence_train", "sub_source": "UNKNOWN_DIFFICULTY" }
UNKNOWN_DIFFICULTY
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "In $\\Delta ABC$, if $(b+c)(b-c)=a(a-c)$, then $\\angle B=$ (    )\nA: ${60}^{\\circ}$\nB: ${90}^{\\circ}$\nC: ${120}^{\\circ}$\nD: ${150}^{\\circ}$", "role": "user" } ]
math
{ "ground_truth": "A", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-1458", "rating": null, "split": "confidence_train", "sub_source": "cn_k12" }
cn_k12
codecontests
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Let's call a string adorable if its letters can be r...
code
{ "ground_truth": "{\"inputs\": [\"yeee\\n\", \"zzcxx\\n\", \"ababa\\n\", \"a\\n\", \"abbbc\\n\", \"aa\\n\", \"abcc\\n\", \"adb\\n\", \"aabcdef\\n\", \"ssab\\n\", \"aaabbbccc\\n\", \"bbbba\\n\", \"aaaaaaabbbbbbcder\\n\", \"abca\\n\", \"aabbcc\\n\", \"abc\\n\", \"abbc\\n\", \"ccdd\\n\", \"aabbccddee\\n\", \"abcdaaaa\\...
{ "difficulty": null, "id": "955_B. Not simply beatiful strings", "rating": 1400, "split": "confidence_train", "sub_source": "rating_1400" }
rating_1400
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Due to atheistic Soviet past, Christmas wasn't offic...
code
{ "ground_truth": "{\"inputs\": [\"1 1\\n1 1000\\n\", \"1 1\\n1000 1000\\n\", \"1 1\\n1000 1\\n\", \"10 10\\n7 5 5 5 9 10 9 8 7 5 10\\n4 2 8 2 9 1 2 8 10 7 10\\n9 7 7 2 5 1 5 4 7 9 7\\n7 3 2 10 6 9 10 2 4 2 4\\n1 4 8 6 9 2 1 3 6 2 8\\n2 4 10 7 1 1 7 9 8 9 8\\n8 3 10 9 4 9 9 1 9 6 3\\n8 10 7 2 6 2 1 3 9 7 5\\n1 8 3 1 ...
{ "difficulty": "VERY_HARD", "id": "taco-cft-VERY_HARD-1530", "rating": null, "split": "confidence_train", "sub_source": "VERY_HARD" }
VERY_HARD
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "Calculate:<br/>$(1)\\left(+17\\right)+\\left(-12\\right)$;<br/>$(2)-\\frac{1}{7}-(-\\frac{6}{7})$.", "role": "user" } ]
math
{ "ground_truth": "\\frac{5}{7}", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-757", "rating": null, "split": "confidence_train", "sub_source": "cn_k12" }
cn_k12
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "Grandpa's age this year is 6 times the age of his grandson. After 4 years, the sum of the ages of the grandfather and grandson will be 78. How old is the grandson this year?", ...
math
{ "ground_truth": "10", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-2793", "rating": null, "split": "confidence_train", "sub_source": "orca_math" }
orca_math
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "You are given two set of points. The first set is de...
code
{ "ground_truth": "{\"inputs\": [\"1 -1 1\\n0 0 0\\n\", \"-1 0 0\\n-2 1 2\\n\", \"-1 -1 1\\n1 1 0\\n\", \"2 2 1\\n1 2 0\\n\", \"-1 1 1\\n0 1 1\\n\", \"0 -1 -1\\n-2 -2 -1\\n\", \"-2 -1 1\\n-1 1 1\\n\", \"1 0 0\\n0 1 0\\n\", \"1 1 -1\\n-1 0 -1\\n\", \"-2 -1 0\\n-2 2 2\\n\", \"1 1 -1\\n0 1 0\\n\", \"0 2 -2\\n-2 2 0\\n\"...
{ "difficulty": "HARD", "id": "taco-cft-HARD-2605", "rating": null, "split": "confidence_train", "sub_source": "HARD" }
HARD
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "{ ( 476 + 424 ) 2 - 4 x 476 x 424 } = ? What is the result of this calculation?", "role": "user" } ]
math
{ "ground_truth": "4624", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-1731", "rating": null, "split": "confidence_train", "sub_source": "orca_math" }
orca_math
codecontests
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "One day Igor K. stopped programming and took up math...
code
{ "ground_truth": "{\"inputs\": [\"1\\n3 5\\n\", \"2\\n5 3\\n1 3\\n\", \"5\\n1 3\\n1 4\\n2 1\\n4 3\\n1 5\\n\", \"3\\n4 1\\n4 5\\n2 1\\n\", \"5\\n1 5\\n3 4\\n1 4\\n5 4\\n4 2\\n\", \"6\\n3 2\\n2 4\\n3 1\\n3 5\\n5 2\\n1 2\\n\", \"5\\n4 1\\n5 1\\n2 3\\n2 5\\n1 2\\n\", \"1\\n5 4\\n\", \"5\\n3 5\\n4 2\\n1 4\\n3 4\\n5 2\\n\...
{ "difficulty": null, "id": "94_B. Friends", "rating": 1300, "split": "confidence_train", "sub_source": "rating_1300" }
rating_1300
codeforces
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Vasya thinks that lucky tickets are the tickets whos...
code
{ "ground_truth": "{\"inputs\": [\"3\\n123 123 99\", \"6\\n1 1 1 23 10 3\", \"3\\n43440907 58238452 82582355\", \"4\\n31450303 81222872 67526764 17516401\", \"5\\n83280 20492640 21552119 7655071 47966344\", \"6\\n94861402 89285133 30745405 41537407 90189008 83594323\", \"7\\n95136773 99982752 97528336 79027944 968474...
{ "difficulty": null, "id": "43C", "rating": 1300, "split": "confidence_train", "sub_source": "rating_1300" }
rating_1300
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "The new season of the Bytelandian Premier League (BP...
code
{ "ground_truth": "{\"inputs\": [\"3\\n3\\n0 0 0 \\n1 0 1 \\n1 0 0 \\n4\\n0 1 1 0 \\n0 0 2 0 \\n0 2 0 0 \\n1 1 1 0 \\n5\\n0 2 2 1 0 \\n2 0 1 1 0 \\n2 0 0 1 0 \\n0 0 0 0 1 \\n1 1 1 0 0\"], \"outputs\": [\"010\\n0001\\n11001\"]}", "style": "rule" }
{ "difficulty": "MEDIUM_HARD", "id": "taco-cft-MEDIUM_HARD-2056", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM_HARD" }
MEDIUM_HARD
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "Participants in a chess tournament played one game with each other. For each participant \\( A \\), the number of points they earned was calculated (1 point for a win, \\( \\frac...
math
{ "ground_truth": "\\text{It is impossible for all participants to have either all positive or all negative strength coefficients.}", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-80", "rating": null, "split": "confidence_train", "sub_source": "olympiads" }
olympiads
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). Implement the required function(s) as specified; do not read from standard input.", "role": "system" }, { "content": "Ever heard about Dijkstra's shallowest path algorithm...
code
{ "ground_truth": "{\"fn_name\": \"shallowest_path\", \"inputs\": [], \"outputs\": []}", "style": "rule" }
{ "difficulty": "MEDIUM", "id": "taco-cft-MEDIUM-2419", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM" }
MEDIUM
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "The entire network is under the inspection and direc...
code
{ "ground_truth": "{\"inputs\": [[\"techtrishna online event\", \"onlin\"]], \"outputs\": [[\"Y\"]]}", "style": "rule" }
{ "difficulty": "UNKNOWN_DIFFICULTY", "id": "taco-cft-UNKNOWN_DIFFICULTY-3222", "rating": null, "split": "confidence_train", "sub_source": "UNKNOWN_DIFFICULTY" }
UNKNOWN_DIFFICULTY
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "Ten people are sitting around a circular table, each with a fair 6-sided die. Each person rolls their die and stands if they roll a 6, and remains seated otherwise. What is the p...
math
{ "ground_truth": "\\textbf{(A)} \\frac{123}{1024}", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-664", "rating": null, "split": "confidence_train", "sub_source": "synthetic_amc" }
synthetic_amc
codecontests
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "There are n cities in the country where the Old Peyk...
code
{ "ground_truth": "{\"inputs\": [\"4 6\\n1 2 5 2\\n2 3 3 4\\n\", \"2 3\\n5 6\\n5 5\\n\", \"43 5\\n6 7 15 12 15 7 22 33 38 15 7 23 31 21 26 41 25 14 26 33 5 28 22 6 35 17 19 32 41 27 20 25 5 32 37 19 40 9 25 22 10 24 9\\n3 5 3 6 5 4 5 3 3 3 3 6 6 3 3 3 3 3 3 3 3 6 3 3 4 3 4 3 6 4 3 6 3 4 6 3 4 5 4 4 3 3 5\\n\", \"24 3...
{ "difficulty": null, "id": "241_A. Old Peykan", "rating": 1300, "split": "confidence_train", "sub_source": "rating_1300" }
rating_1300
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Disclaimer: there are lots of untranslateable puns i...
code
{ "ground_truth": "{\"inputs\": [\"228\\n1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 31 1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 63 1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 31 1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 127 1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 31 1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 63 1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 31 1 3 1 7 1 3 1 15 1 3 1 7 1 3 1 255 1 3...
{ "difficulty": "HARD", "id": "taco-cft-HARD-1456", "rating": null, "split": "confidence_train", "sub_source": "HARD" }
HARD
codecontests
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "You are given a string. Reverse its characters.\n\nI...
code
{ "ground_truth": "{\"inputs\": [\"secrofedoc\\n\", \"!ssalg-gnikool5\\n\"], \"outputs\": [\"codeforces\\n\", \"5looking-glass!\\n\"]}", "style": "rule" }
{ "difficulty": null, "id": "130_B. Gnikool Ssalg", "rating": 1400, "split": "confidence_train", "sub_source": "rating_1400" }
rating_1400
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "This is the hard version of this problem. In this ve...
code
{ "ground_truth": "{\"inputs\": [\"6\\n5 8 9\\n01001\\n00101\\n6 2 11\\n000001\\n100000\\n5 7 2\\n01000\\n11011\\n7 8 3\\n0111001\\n0100001\\n6 3 4\\n010001\\n101000\\n5 10 1\\n01100\\n01100\\n\"], \"outputs\": [\"8\\n10\\n-1\\n6\\n7\\n0\\n\"]}", "style": "rule" }
{ "difficulty": "HARD", "id": "taco-cft-HARD-1523", "rating": null, "split": "confidence_train", "sub_source": "HARD" }
HARD
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "In the expansion of $({x+\\frac{2}{{\\sqrt{x}}}})^6$, the constant term is ______ (answer with a number).", "role": "user" } ]
math
{ "ground_truth": "240", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-398", "rating": null, "split": "confidence_train", "sub_source": "cn_k12" }
cn_k12
codecontests
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Vasya has found a strange device. On the front panel...
code
{ "ground_truth": "{\"inputs\": [\"4 6\\n\", \"10 1\\n\", \"5001 10000\\n\", \"9097 9998\\n\", \"2 100\\n\", \"886 9383\\n\", \"4000 7997\\n\", \"1 8193\\n\", \"9912 1023\\n\", \"9967 9973\\n\", \"3 6\\n\", \"9102 9103\\n\", \"1918 10000\\n\", \"1 9999\\n\", \"1 8192\\n\", \"10 8722\\n\", \"100 99\\n\", \"99 100\\n\"...
{ "difficulty": null, "id": "520_B. Two Buttons", "rating": 1400, "split": "confidence_train", "sub_source": "rating_1400" }
rating_1400
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Read problems statements in Mandarin Chinese and Ru...
code
{ "ground_truth": "{\"inputs\": [\"3\\nqwrtyuipasdfgjkzxcvbnmheol\\ndummy!\\nbfgjklmopqrstuwxzhvnicdyea\\nabcd b efgd hbi!\\nqwrtyuipasdfgjkzxcvbnmheol\\nDummy!\", \"3\\nqwrtyuipasdfgjkzxcvbnmheol\\ndummy!\\nbfgjklmopqrstuwxzhvnicdyea\\nabcd b efgd hbi!\\nqwrtyuipasdfgjkzxcvbnmheol\\nDummy!\"], \"outputs\": [\"hello!...
{ "difficulty": "MEDIUM_HARD", "id": "taco-cft-MEDIUM_HARD-1324", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM_HARD" }
MEDIUM_HARD
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "A box contains chips of three colors: red, white, and blue. The number of red chips is at least twice the number of white chips, and at most two thirds the number of blue chips. ...
math
{ "ground_truth": "C) 72", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-4147", "rating": null, "split": "confidence_train", "sub_source": "synthetic_amc" }
synthetic_amc
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "If $m$ is the geometric mean of 5 and $\\frac{16}{5}$, then the eccentricity of the conic section $\\frac{x^2}{m} + y^2 = 1$ is\nA: $\\frac{\\sqrt{3}}{2}$\nB: $\\sqrt{5}$\nC: $\\...
math
{ "ground_truth": "\\text{D}", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-5986", "rating": null, "split": "confidence_train", "sub_source": "cn_k12" }
cn_k12
codecontests
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "Petya is preparing for his birthday. He decided that...
code
{ "ground_truth": "{\"inputs\": [\"4\\n4\\n3 7 4 5\\n2 1 2 4\\n4\\n1 2 3 4\\n3 3 3 3\\n2\\n1 2\\n10 10\\n2\\n10 10\\n1 2\\n\", \"4\\n4\\n3 7 4 5\\n2 1 2 4\\n4\\n1 2 3 4\\n3 3 3 5\\n2\\n1 2\\n10 10\\n2\\n10 10\\n1 2\\n\", \"4\\n4\\n3 7 4 2\\n2 1 2 4\\n4\\n1 2 3 4\\n3 3 3 3\\n2\\n1 2\\n10 10\\n2\\n10 10\\n1 2\\n\", \"4...
{ "difficulty": null, "id": "1443_C. The Delivery Dilemma", "rating": 1400, "split": "confidence_train", "sub_source": "rating_1400" }
rating_1400
math_box
[ { "content": "Please reason step by step, and put your final answer within \\boxed{}.", "role": "system" }, { "content": "One tourist covers a distance of 20 km 2.5 hours faster than another. If the first tourist were to reduce their speed by 2 km/h, and the second were to increase their speed by 50...
math
{ "ground_truth": "8 \\text{ km/h for the first tourist, 4 \\text{ km/h for the second tourist}}", "style": "rule" }
{ "difficulty": null, "id": "numina-cft-2143", "rating": null, "split": "confidence_train", "sub_source": "olympiads" }
olympiads
taco
[ { "content": "Please reason step by step, then provide your complete solution as a single Python code block (```python ... ```). The program must read from standard input and write the answer to standard output.", "role": "system" }, { "content": "These days, many boys use beautiful girls' photos as...
code
{ "ground_truth": "{\"inputs\": [\"rahul\", \"palak\", \"payal\", \"anisha\"], \"outputs\": [\"IGNORE HIM!\", \"CHAT WITH HER!\", \"CHAT WITH HER!\", \"IGNORE HIM!\"]}", "style": "rule" }
{ "difficulty": "UNKNOWN_DIFFICULTY", "id": "taco-cft-UNKNOWN_DIFFICULTY-2200", "rating": null, "split": "confidence_train", "sub_source": "UNKNOWN_DIFFICULTY" }
UNKNOWN_DIFFICULTY
End of preview. Expand in Data Studio

rl-value-confidence-train

A held-out training set for a confidence / correctness estimator (~5.7k prompts), paired with YangyiYY/rl-value-eval (evaluation) and the nct-ppo actor/critic models. Each row is a (prompt, ground_truth) in the exact verl RL schema, so the same rule-based verifiers score it (math: \boxed{} grading; code: stdin/stdout or function test cases).

Built to be disjoint from both the RL training data and the calibration eval sets (enforced by md5 of the normalized problem text — 0 overlap), at matched difficulty where the pools allow.

Composition (5728 rows)

data_source n source difficulty
math_box (numina) 2000 AI-MO/NuminaMath-CoT train, boxed answers same distribution as RL training (sub-datasets: cn_k12, synthetic_math, olympiads, orca_math, …)
taco 2998 BAAI/TACO train sampled to match the RL-training taco difficulty mix (EASY/MEDIUM/HARD/…)
codecontests 655 deepmind/code_contests (train+valid+test) cf_rating 1201–1400
codeforces 75 MatrixStudio/Codeforces-Python-Submissions rating 1201–1400

Note on code difficulty: the RL training + eval sets consumed all Codeforces/CodeContests problems at rating ≤1200 (the training band). This held-out set therefore takes the closest available band (1201–1400) for code — one notch above training — since ≤1200 has zero disjoint problems left. numina and taco are at matched training difficulty.

Schema

data_source · prompt (list of {role, content} chat messages) · ability (math/code) · reward_model.ground_truth (rule verifier target) · extra_info (id, rating, difficulty, sub_source, split="confidence_train").

Intended use

Train a correctness/confidence estimator (e.g. a value head or probe) on rollouts from the frozen nct-ppo actor, then evaluate calibration (ECE/AUROC) on the disjoint rl-value-eval set.

Downloads last month
-