Datasets:

YangyiYY
/

rl-value-confidence-train-code

data_source stringclasses 3 values	prompt listlengths 2 2	ability stringclasses 1 value	reward_model dict	sub_source stringclasses 12 values	extra_info dict
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": "The problem describes the properties of a command li...	code	{ "ground_truth": "{\"inputs\": [\"\\\"RUn.exe O\\\" \\\"\\\" \\\" 2ne, \\\" two! . \\\" \\\"\\n\", \"firstarg second \\\"\\\" \\n\", \"\\\" \\\" \\n\", \"j \\n\", \"B\\n\", \"\\\"7\\\" \\\"W \\\" \\\"\\\" \\\"\\\" \\\"a \\\" \\\"\\\" \\\"\\\" \\\"\\\" y \\n\", \"A\\n\", \"\\\"\\\"\\n\", \"\\\"RUn.exe O...	rating_1201-1400	{ "id": "291_B. Command Line Arguments", "rating": 1300, "split": "confidence_train", "sub_source": "rating_1201-1400" }
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 integers, A_1, A_2, ..., A_N, arranged i...	code	{ "ground_truth": "{\"inputs\": [\"11\\n-1000000000 1000000000 -1000000000 1000000000 -1000000000 0 1000000000 -1000000000 1000000000 -1000000000 1000000000\", \"5\\n10 -4 -8 -11 3\", \"3\\n-10 5 -4\", \"11\\n-1000000000 1000000000 -364513171 1000000000 -1000000000 0 1000000000 -1000000000 1000000000 -1000000000 1000...	none	{ "id": "p03062 AtCoder Beginner Contest 125 - Flipping Signs", "rating": 0, "split": "confidence_train", "sub_source": "none" }
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": "You are given a rectangular cake, represented as an ...	code	{ "ground_truth": "{\"inputs\": [\"3 4\\nS...\\n....\\n..S.\", \"2 2\\n..\\n..\", \"2 2\\nSS\\nSS\", \"7 3\\nS..\\nS..\\nS..\\nS..\\nS..\\nS..\\nS..\", \"3 5\\n..S..\\nSSSSS\\n..S..\", \"10 10\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSSSSSS\\nSSSSSS...	rating_<=1200	{ "id": "330A", "rating": 800, "split": "confidence_train", "sub_source": "rating_<=1200" }
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": "The only difference between easy and hard versions i...	code	{ "ground_truth": "{\"inputs\": [\"6\\n8\\n1 1 2 2 3 2 1 1\\n3\\n1 3 3\\n4\\n1 10 10 1\\n1\\n26\\n2\\n2 1\\n3\\n1 1 1\\n\", \"6\\n8\\n1 1 2 2 3 2 1 1\\n3\\n1 3 3\\n4\\n1 10 10 1\\n1\\n26\\n2\\n2 2\\n3\\n1 1 1\\n\", \"6\\n8\\n1 1 2 2 3 2 1 1\\n3\\n1 3 2\\n4\\n1 10 10 1\\n1\\n26\\n2\\n2 1\\n3\\n1 1 1\\n\", \"6\\n8\\n1 ...	rating_1601-1900	{ "id": "1335_E2. Three Blocks Palindrome (hard version)", "rating": 1800, "split": "confidence_train", "sub_source": "rating_1601-1900" }
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 an array A with size N (indexed from 0...	code	{ "ground_truth": "{\"inputs\": [[\"2\", \"2 3\", \"1 2\", \"3 2\", \"1 -2 1\", \"\", \"\"]], \"outputs\": [[\"9\", \"2\"]]}", "style": "rule" }	MEDIUM_HARD	{ "id": "taco-train-8115", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM_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": "Let's consider a rectangular table R consisting of N...	code	{ "ground_truth": "{\"inputs\": [[\"3\", \"1 1\", \"2 2\", \"2 3\"]], \"outputs\": [[\"2\", \"25\", \"273\"]]}", "style": "rule" }	MEDIUM_HARD	{ "id": "taco-train-2090", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM_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": "You are given a rooted tree with N nodes and the roo...	code	{ "ground_truth": "{\"inputs\": [\"7 7 1\\n1 2\\n2 3\\n2 4\\n2 5\\n5 6\\n6 7\\nU 5 10 2\\nU 4 5 3\\nQ 1 7\\nU 6 7 4\\nQ 2 7\\nQ 1 4\\nQ 2 4\\n\"], \"outputs\": [\"36\\n54\\n5\\n5\\n\"]}", "style": "rule" }	MEDIUM_HARD	{ "id": "taco-train-2029", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM_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": "Palindromic characteristics of string s with length ...	code	{ "ground_truth": "{\"inputs\": [\"abacaba\\n\", \"abba\\n\", \"qqqpvmgd\\n\", \"wyemcafatp\\n\", \"qqqpvngd\\n\", \"wyempafatc\\n\", \"abacbaa\\n\", \"abca\\n\", \"wyempaf`tc\\n\", \"acca\\n\", \"aaacaba\\n\", \"qqgvpoqd\\n\", \"`aadaba\\n\", \"acad\\n\", \"qqgvqnqd\\n\", \"aacc\\n\", \"rqfnqwpe\\n\", \"qqqvpngd\\n\...	rating_1601-1900	{ "id": "835_D. Palindromic characteristics", "rating": 1900, "split": "confidence_train", "sub_source": "rating_1601-1900" }
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": "Given a grid of dimension nxm where each cell in the ...	code	{ "ground_truth": "{\"inputs\": [\"grid = {{0,1,2},{0,1,2},{2,1,1}}\", \"grid = {{2,2,0,1}}\"], \"outputs\": [\"1\", \"-1\"]}", "style": "rule" }	MEDIUM	{ "id": "taco-train-9517", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM" }
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": "Masha really loves algebra. On the last lesson, her ...	code	{ "ground_truth": "{\"inputs\": [\"123 1 2143435 4\\n123 11 -5453 141245\\n\", \"3 2 30 4\\n6 14 25 48\\n\", \"123 1 2143435 4\\n54343 -13 6 124\\n\", \"-100 0 50 1\\n0\\n\", \"-1000 0 10 1\\n5\\n\", \"-3 1 100 1\\n3\\n\", \"1000000000 100000 1000000000 4\\n5433 13 6 0\\n\", \"5 1 5 5\\n1 2 3 4 0\\n\", \"2 -1 6 1\\n2...	rating_1601-1900	{ "id": "789_B. Masha and geometric depression", "rating": 1700, "split": "confidence_train", "sub_source": "rating_1601-1900" }
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": "write me a function `stringy` that takes a `size` and...	code	{ "ground_truth": "{\"fn_name\": \"stringy\", \"inputs\": [[3], [5], [12], [26], [28]], \"outputs\": [[\"101\"], [\"10101\"], [\"101010101010\"], [\"10101010101010101010101010\"], [\"1010101010101010101010101010\"]]}", "style": "rule" }	EASY	{ "id": "taco-train-1560", "rating": null, "split": "confidence_train", "sub_source": "EASY" }
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": "Manao has invented a new mathematical term — a beaut...	code	{ "ground_truth": "{\"inputs\": [\"2 2\\n\", \"4 3\\n\", \"21 21\\n\", \"10 1\\n\", \"4 4\\n\", \"1 1\\n\", \"5 5\\n\", \"100 11\\n\", \"16 55\\n\", \"13 71\\n\", \"30 40\\n\", \"6 6\\n\", \"37 42\\n\", \"4 6\\n\", \"1 4\\n\", \"2 10\\n\", \"6 7\\n\", \"28 28\\n\", \"10 6\\n\", \"9 6\\n\", \"2 6\\n\", \"6 4\\n\", \"6...	MEDIUM	{ "id": "taco-test-119", "rating": null, "split": "confidence_train", "sub_source": "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": "Vasya has a grid with $2$ rows and $n$ columns. He c...	code	{ "ground_truth": "{\"inputs\": [\"6\\n2\\nRG\\nRB\\n4\\nGRBG\\nGBGB\\n5\\nGGGGG\\nBBBBB\\n7\\nBBBBBBB\\nRRRRRRR\\n8\\nRGBRRGBR\\nRGGRRBGR\\n1\\nG\\nG\\n\", \"1\\n7\\nGGGRRBR\\nGGGRRBB\\n\", \"1\\n55\\nBBBBBBBBBBBBBBBBBBBGGGGGGGGGGGGGGGGGGGGGGGGGGGGBBBBGBGB\\nBBBBBBBBBBBBBBBBBBBGGGGGGGGGGGGGGGGGGGGGGGGGGGGBBBBGBGB\\n...	EASY	{ "id": "taco-train-3511", "rating": null, "split": "confidence_train", "sub_source": "EASY" }
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 an array $a$ consisting of $n$ positiv...	code	{ "ground_truth": "{\"inputs\": [\"5\\n4 3\\n1 2 1 3\\n10 6\\n8 7 1 8 3 7 5 10 8 9\\n5 10\\n20 100 50 20 100500\\n10 25\\n24 24 24 24 24 24 24 24 24 24\\n8 8\\n1 2 3 4 5 6 7 8\\n\", \"1\\n1 1000000000\\n99999999\\n\", \"1\\n5 3\\n20 100 50 20 1005002\\n\", \"1\\n1 1000000000\\n99999999\\n\", \"1\\n5 3\\n20 100 50 20 ...	MEDIUM	{ "id": "taco-train-5235", "rating": null, "split": "confidence_train", "sub_source": "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 Bubble Cup hypothesis stood unsolved for $130$ y...	code	{ "ground_truth": "{\"inputs\": [\"2\\n2 4\\n\", \"1\\n9\\n\", \"5\\n4 1 8 3 9\\n\", \"6\\n8 7 8 6 8 9\\n\", \"8\\n1 1 7 6 1 5 8 7\\n\", \"7\\n9 6 3 1 3 1 7\\n\", \"3\\n9 2 8\\n\", \"5\\n3 7 3 4 7\\n\", \"5\\n4 8 3 2 6\\n\", \"5\\n2 7 4 8 3\\n\", \"5\\n3 7 3 4 7\\n\", \"100\\n751 557 779 329 39 213 844 289 137 50 951...	HARD	{ "id": "taco-train-10251", "rating": null, "split": "confidence_train", "sub_source": "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 directed graph G with n vertices and...	code	{ "ground_truth": "{\"inputs\": [\"5 2\\n1 4\\n2 5\\n\", \"4 5\\n1 2\\n2 3\\n3 1\\n2 4\\n4 1\\n\", \"4 4\\n1 1\\n1 2\\n2 1\\n1 2\\n\", \"4 4\\n1 2\\n2 1\\n3 4\\n4 3\\n\", \"3 4\\n1 2\\n2 1\\n1 3\\n3 1\\n\", \"6 6\\n1 2\\n2 3\\n3 4\\n4 5\\n1 6\\n6 5\\n\", \"3 3\\n1 2\\n3 1\\n3 2\\n\", \"2 1\\n2 1\\n\", \"2 2\\n1 2\\n2...	rating_>1900	{ "id": "183_C. Cyclic Coloring", "rating": 2200, "split": "confidence_train", "sub_source": "rating_>1900" }
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": "Given a number N, find the N^{th} term in the series ...	code	{ "ground_truth": "{\"inputs\": [\"N = 4\", \"N = 3\"], \"outputs\": [\"10\", \"6\"]}", "style": "rule" }	EASY	{ "id": "taco-train-6716", "rating": null, "split": "confidence_train", "sub_source": "EASY" }
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 computing, the collection of four bits is called ...	code	{ "ground_truth": "{\"inputs\": [\"4\\n8\\n17\\n52\\n3\"], \"outputs\": [\"Good\\nNot Good\\nGood\\nNot Good\"]}", "style": "rule" }	EASY	{ "id": "taco-train-11057", "rating": null, "split": "confidence_train", "sub_source": "EASY" }
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": "Dilpreet wants to paint his dog's home that has n boa...	code	{ "ground_truth": "{\"inputs\": [\"n = 5\\nk = 3\\narr[] = {5,10,30,20,15}\", \"n = 4\\nk = 2\\narr[] = {10,20,30,40}\"], \"outputs\": [\"35\", \"60\"]}", "style": "rule" }	MEDIUM_HARD	{ "id": "taco-train-7", "rating": null, "split": "confidence_train", "sub_source": "MEDIUM_HARD" }
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": "Write an algorithm that takes an array and moves all ...	code	{ "ground_truth": "{\"fn_name\": \"move_zeros\", \"inputs\": [[[1, 2, 0, 1, 0, 1, 0, 3, 0, 1]], [[9, 0.0, 0, 9, 1, 2, 0, 1, 0, 1, 0.0, 3, 0, 1, 9, 0, 0, 0, 0, 9]], [[\"a\", 0, 0, \"b\", \"c\", \"d\", 0, 1, 0, 1, 0, 3, 0, 1, 9, 0, 0, 0, 0, 9]], [[\"a\", 0, 0, \"b\", null, \"c\", \"d\", 0, 1, false, 0, 1, 0, 3, [], 0, ...	EASY	{ "id": "taco-train-413", "rating": null, "split": "confidence_train", "sub_source": "EASY" }
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": "Chain Disappearance Puzzle\n\nWe are playing a puzzl...	code	{ "ground_truth": "{\"inputs\": [\"1\\n6 9 9 9 9\\n5\\n5 9 5 5 9\\n5 5 6 9 9\\n4 6 3 6 9\\n3 3 2 9 9\\n2 2 1 1 1\\n10\\n3 5 6 5 6\\n2 2 2 8 3\\n6 2 5 9 2\\n7 7 7 6 1\\n4 6 6 4 9\\n8 9 1 1 8\\n5 6 1 8 1\\n6 8 2 1 2\\n9 6 3 3 5\\n5 3 8 8 8\\n5\\n1 2 3 4 5\\n6 7 8 9 1\\n2 3 4 5 6\\n7 8 9 1 2\\n3 4 5 6 7\\n3\\n2 2 8 7 4\...	none	{ "id": "p00774 Chain Disappearance Puzzle", "rating": 0, "split": "confidence_train", "sub_source": "none" }
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 piles of stones of sizes a1, a2, ..., an...	code	{ "ground_truth": "{\"inputs\": [\"5\\n2 3 4 1 1\\n2\\n2 3\\n\", \"3\\n1 6 8\\n1\\n6\\n\", \"1\\n10\\n5\\n5 3 7 7 1\\n\", \"4\\n7 4 1 7\\n3\\n6 8 3\\n\", \"1\\n1\\n3\\n2 1 10\\n\", \"1\\n1\\n3\\n2 1 10\\n\", \"2\\n2 9\\n5\\n4 10 7 3 4\\n\", \"1\\n10\\n5\\n5 3 7 7 1\\n\", \"1\\n2\\n5\\n7 3 9 8 1\\n\", \"2\\n9 3\\n1\\n...	rating_1601-1900	{ "id": "227_D. Naughty Stone Piles", "rating": 1900, "split": "confidence_train", "sub_source": "rating_1601-1900" }

End of preview. Expand in Data Studio

rl-value-confidence-train-code

Code-domain training set (~4.9K) for a confidence / correctness estimator, sampled from the raw competitive-programming datasets Codeforces (MatrixStudio), BAAI/TACO, and deepmind/code_contests. Pairs with YangyiYY/rl-value-eval-code (eval) and a ~25K RL-training split from the same pool (disjoint).

Built by taking ALL distinct usable problems, deduped by problem text GLOBALLY (these datasets heavily re-share Codeforces problems; small-source-first priority cf→cc→taco), split per source 50:10:5 into train/conf/eval (0 problem-text overlap across splits).

Grader / ground_truth

data_source is codeforces / taco / codecontests — each routes to the shared reference grader (reference/, apps_execution) both for the RL reward and these labels, so they are IDENTICAL. reward_model.ground_truth = input_output JSON with the full hidden test suite, capped at 50 test cases/problem (problems whose gt still exceeds 1MB — a single giant test — are dropped, as they can't be graded within the 60s/problem cap).

Schema

data_source · prompt=[{system}, {user: problem}] · ability="code" · reward_model.ground_truth · sub_source (codeforces/codecontests → rating band; taco → difficulty) · extra_info (id, rating, sub_source).

Downloads last month: 26