prompt listlengths 2 2 | label dict | metadata dict |
|---|---|---|
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14or28",
"budget": 4287
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003847",
"normalized_example_id": "ex_003847",
"source_idx": 3847,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7145,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14or28",
"budget": 5716
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003847",
"normalized_example_id": "ex_003847",
"source_idx": 3847,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7145,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "14or28",
"budget": 7145
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003847",
"normalized_example_id": "ex_003847",
"source_idx": 3847,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7145,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1 + \\sqrt{5}}{2}",
"budget": 3657
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003848",
"normalized_example_id": "ex_003848",
"source_idx": 3848,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6095,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1 + \\sqrt{5}}{2}",
"budget": 4876
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003848",
"normalized_example_id": "ex_003848",
"source_idx": 3848,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6095,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1 + \\sqrt{5}}{2}",
"budget": 6095
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003848",
"normalized_example_id": "ex_003848",
"source_idx": 3848,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6095,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{8}",
"budget": 3758
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003849",
"normalized_example_id": "ex_003849",
"source_idx": 3849,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6263,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{8}",
"budget": 5010
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003849",
"normalized_example_id": "ex_003849",
"source_idx": 3849,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6263,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{1}{8}",
"budget": 6263
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003849",
"normalized_example_id": "ex_003849",
"source_idx": 3849,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6263,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1333",
"budget": 4270
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003850",
"normalized_example_id": "ex_003850",
"source_idx": 3850,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7117,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1333",
"budget": 5694
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003850",
"normalized_example_id": "ex_003850",
"source_idx": 3850,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7117,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1333",
"budget": 7117
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003850",
"normalized_example_id": "ex_003850",
"source_idx": 3850,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7117,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1003",
"budget": 3605
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003851",
"normalized_example_id": "ex_003851",
"source_idx": 3851,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6008,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1003",
"budget": 4806
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003851",
"normalized_example_id": "ex_003851",
"source_idx": 3851,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6008,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1003",
"budget": 6008
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003851",
"normalized_example_id": "ex_003851",
"source_idx": 3851,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6008,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "x = 0 \\text{ or } 1",
"budget": 5234
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003860",
"normalized_example_id": "ex_003860",
"source_idx": 3860,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8724,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "x = 0 \\text{ or } 1",
"budget": 6979
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003860",
"normalized_example_id": "ex_003860",
"source_idx": 3860,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8724,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "x = 0 \\text{ or } 1",
"budget": 8724
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003860",
"normalized_example_id": "ex_003860",
"source_idx": 3860,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8724,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{\\sqrt{3}}{8}",
"budget": 1335
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003861",
"normalized_example_id": "ex_003861",
"source_idx": 3861,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2225,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{\\sqrt{3}}{8}",
"budget": 1780
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003861",
"normalized_example_id": "ex_003861",
"source_idx": 3861,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2225,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{\\sqrt{3}}{8}",
"budget": 2225
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003861",
"normalized_example_id": "ex_003861",
"source_idx": 3861,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2225,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(4,9)",
"budget": 740
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003864",
"normalized_example_id": "ex_003864",
"source_idx": 3864,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1234,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(4,9)",
"budget": 987
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003864",
"normalized_example_id": "ex_003864",
"source_idx": 3864,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1234,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "(4,9)",
"budget": 1234
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003864",
"normalized_example_id": "ex_003864",
"source_idx": 3864,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 1234,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 4036
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003866",
"normalized_example_id": "ex_003866",
"source_idx": 3866,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6726,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 5381
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003866",
"normalized_example_id": "ex_003866",
"source_idx": 3866,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6726,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "0",
"budget": 6726
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003866",
"normalized_example_id": "ex_003866",
"source_idx": 3866,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6726,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "AC=48,BC=36",
"budget": 4415
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003869",
"normalized_example_id": "ex_003869",
"source_idx": 3869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7359,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "AC=48,BC=36",
"budget": 5887
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003869",
"normalized_example_id": "ex_003869",
"source_idx": 3869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7359,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "AC=48,BC=36",
"budget": 7359
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003869",
"normalized_example_id": "ex_003869",
"source_idx": 3869,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7359,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "224",
"budget": 1928
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003874",
"normalized_example_id": "ex_003874",
"source_idx": 3874,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3213,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "224",
"budget": 2570
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003874",
"normalized_example_id": "ex_003874",
"source_idx": 3874,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3213,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "224",
"budget": 3213
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003874",
"normalized_example_id": "ex_003874",
"source_idx": 3874,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3213,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2916",
"budget": 5383
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003875",
"normalized_example_id": "ex_003875",
"source_idx": 3875,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8971,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2916",
"budget": 7177
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003875",
"normalized_example_id": "ex_003875",
"source_idx": 3875,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8971,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "2916",
"budget": 8971
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003875",
"normalized_example_id": "ex_003875",
"source_idx": 3875,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8971,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\binom{21}{9}\\cdot2^{-\\frac{1}{2}}x^{\\frac{5}{2}}y^{\\frac{5}{2}}",
"budget": 4798
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003876",
"normalized_example_id": "ex_003876",
"source_idx": 3876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7997,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\binom{21}{9}\\cdot2^{-\\frac{1}{2}}x^{\\frac{5}{2}}y^{\\frac{5}{2}}",
"budget": 6398
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003876",
"normalized_example_id": "ex_003876",
"source_idx": 3876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7997,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "-\\binom{21}{9}\\cdot2^{-\\frac{1}{2}}x^{\\frac{5}{2}}y^{\\frac{5}{2}}",
"budget": 7997
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003876",
"normalized_example_id": "ex_003876",
"source_idx": 3876,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7997,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{x^{2}}{4\\sqrt{3}}",
"budget": 3962
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003880",
"normalized_example_id": "ex_003880",
"source_idx": 3880,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6604,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{x^{2}}{4\\sqrt{3}}",
"budget": 5283
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003880",
"normalized_example_id": "ex_003880",
"source_idx": 3880,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6604,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{x^{2}}{4\\sqrt{3}}",
"budget": 6604
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003880",
"normalized_example_id": "ex_003880",
"source_idx": 3880,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6604,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\left\\lfloor \\frac{n^2}{2} \\right\\rfloor",
"budget": 4701
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003881",
"normalized_example_id": "ex_003881",
"source_idx": 3881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7835,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\left\\lfloor \\frac{n^2}{2} \\right\\rfloor",
"budget": 6268
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003881",
"normalized_example_id": "ex_003881",
"source_idx": 3881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7835,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\left\\lfloor \\frac{n^2}{2} \\right\\rfloor",
"budget": 7835
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003881",
"normalized_example_id": "ex_003881",
"source_idx": 3881,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 7835,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(n)=kn^{2}",
"budget": 6084
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003882",
"normalized_example_id": "ex_003882",
"source_idx": 3882,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10140,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(n)=kn^{2}",
"budget": 8112
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003882",
"normalized_example_id": "ex_003882",
"source_idx": 3882,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10140,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "f(n)=kn^{2}",
"budget": 10140
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003882",
"normalized_example_id": "ex_003882",
"source_idx": 3882,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 10140,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "30",
"budget": 2671
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003883",
"normalized_example_id": "ex_003883",
"source_idx": 3883,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4452,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "30",
"budget": 3562
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003883",
"normalized_example_id": "ex_003883",
"source_idx": 3883,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4452,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "30",
"budget": 4452
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003883",
"normalized_example_id": "ex_003883",
"source_idx": 3883,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4452,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 2126
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003885",
"normalized_example_id": "ex_003885",
"source_idx": 3885,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3544,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 2835
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003885",
"normalized_example_id": "ex_003885",
"source_idx": 3885,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3544,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "18",
"budget": 3544
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003885",
"normalized_example_id": "ex_003885",
"source_idx": 3885,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3544,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2\\sqrt{3}}{3}",
"budget": 5724
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003886",
"normalized_example_id": "ex_003886",
"source_idx": 3886,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9540,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2\\sqrt{3}}{3}",
"budget": 7632
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003886",
"normalized_example_id": "ex_003886",
"source_idx": 3886,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9540,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2\\sqrt{3}}{3}",
"budget": 9540
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003886",
"normalized_example_id": "ex_003886",
"source_idx": 3886,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 9540,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "561",
"budget": 2129
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003891",
"normalized_example_id": "ex_003891",
"source_idx": 3891,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3548,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "561",
"budget": 2838
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003891",
"normalized_example_id": "ex_003891",
"source_idx": 3891,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3548,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "561",
"budget": 3548
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003891",
"normalized_example_id": "ex_003891",
"source_idx": 3891,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3548,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3",
"budget": 4957
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003892",
"normalized_example_id": "ex_003892",
"source_idx": 3892,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8262,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3",
"budget": 6610
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003892",
"normalized_example_id": "ex_003892",
"source_idx": 3892,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8262,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "3",
"budget": 8262
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003892",
"normalized_example_id": "ex_003892",
"source_idx": 3892,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 8262,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\lambda=3",
"budget": 1667
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003894",
"normalized_example_id": "ex_003894",
"source_idx": 3894,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2779,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\lambda=3",
"budget": 2223
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003894",
"normalized_example_id": "ex_003894",
"source_idx": 3894,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2779,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\lambda=3",
"budget": 2779
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003894",
"normalized_example_id": "ex_003894",
"source_idx": 3894,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2779,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2930455}{6^{10}}\\approx0.0485",
"budget": 4166
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003898",
"normalized_example_id": "ex_003898",
"source_idx": 3898,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6944,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2930455}{6^{10}}\\approx0.0485",
"budget": 5555
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003898",
"normalized_example_id": "ex_003898",
"source_idx": 3898,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6944,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\frac{2930455}{6^{10}}\\approx0.0485",
"budget": 6944
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003898",
"normalized_example_id": "ex_003898",
"source_idx": 3898,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 6944,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\sqrt{10}",
"budget": 2963
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003899",
"normalized_example_id": "ex_003899",
"source_idx": 3899,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4939,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\sqrt{10}",
"budget": 3951
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003899",
"normalized_example_id": "ex_003899",
"source_idx": 3899,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4939,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "\\sqrt{10}",
"budget": 4939
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003899",
"normalized_example_id": "ex_003899",
"source_idx": 3899,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4939,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9\\frac{1}{3},9\\frac{2}{3}",
"budget": 1469
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003900",
"normalized_example_id": "ex_003900",
"source_idx": 3900,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2448,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9\\frac{1}{3},9\\frac{2}{3}",
"budget": 1958
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003900",
"normalized_example_id": "ex_003900",
"source_idx": 3900,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2448,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9\\frac{1}{3},9\\frac{2}{3}",
"budget": 2448
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003900",
"normalized_example_id": "ex_003900",
"source_idx": 3900,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2448,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "E",
"budget": 1759
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003901",
"normalized_example_id": "ex_003901",
"source_idx": 3901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2932,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "E",
"budget": 2346
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003901",
"normalized_example_id": "ex_003901",
"source_idx": 3901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2932,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "E",
"budget": 2932
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003901",
"normalized_example_id": "ex_003901",
"source_idx": 3901,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2932,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "588",
"budget": 1498
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003904",
"normalized_example_id": "ex_003904",
"source_idx": 3904,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2496,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "588",
"budget": 1997
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003904",
"normalized_example_id": "ex_003904",
"source_idx": 3904,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2496,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "588",
"budget": 2496
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003904",
"normalized_example_id": "ex_003904",
"source_idx": 3904,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2496,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 1258
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003906",
"normalized_example_id": "ex_003906",
"source_idx": 3906,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2097,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 1678
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003906",
"normalized_example_id": "ex_003906",
"source_idx": 3906,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2097,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "35",
"budget": 2097
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003906",
"normalized_example_id": "ex_003906",
"source_idx": 3906,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2097,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 1278
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003907",
"normalized_example_id": "ex_003907",
"source_idx": 3907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2130,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 1704
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003907",
"normalized_example_id": "ex_003907",
"source_idx": 3907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2130,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "9",
"budget": 2130
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003907",
"normalized_example_id": "ex_003907",
"source_idx": 3907,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2130,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "37",
"budget": 2454
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003909",
"normalized_example_id": "ex_003909",
"source_idx": 3909,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4090,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "37",
"budget": 3272
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003909",
"normalized_example_id": "ex_003909",
"source_idx": 3909,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4090,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "37",
"budget": 4090
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003909",
"normalized_example_id": "ex_003909",
"source_idx": 3909,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4090,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "x",
"budget": 1540
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003913",
"normalized_example_id": "ex_003913",
"source_idx": 3913,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2566,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "x",
"budget": 2053
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003913",
"normalized_example_id": "ex_003913",
"source_idx": 3913,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2566,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "x",
"budget": 2566
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003913",
"normalized_example_id": "ex_003913",
"source_idx": 3913,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 2566,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 2819
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003914",
"normalized_example_id": "ex_003914",
"source_idx": 3914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4698,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 3758
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003914",
"normalized_example_id": "ex_003914",
"source_idx": 3914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4698,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "1",
"budget": 4698
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003914",
"normalized_example_id": "ex_003914",
"source_idx": 3914,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 4698,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "k_{1}+k_{2}=0",
"budget": 1950
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003916",
"normalized_example_id": "ex_003916",
"source_idx": 3916,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3250,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "k_{1}+k_{2}=0",
"budget": 2600
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003916",
"normalized_example_id": "ex_003916",
"source_idx": 3916,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3250,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "k_{1}+k_{2}=0",
"budget": 3250
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003916",
"normalized_example_id": "ex_003916",
"source_idx": 3916,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3250,
... |
[
{
"role": "system",
"content": "You are a reasoning control agent that guides a step-by-step mathematical reasoner.\n\nEach turn you receive the reasoner's latest step and the remaining token budget (as a percentage). On the first turn you receive the question instead. Considering the reasoner's current pro... | {
"answer": "625",
"budget": 2275
} | {
"data_type": "full_prompt_per_prompt_calibrated_budget",
"source_example_id": "ex_003917",
"normalized_example_id": "ex_003917",
"source_idx": 3917,
"augmentation": "online_rl_full_prompt_per_prompt_calibrated_budget",
"budget_source": "per_prompt_calibrated_length_ratios",
"calibrated_length": 3791,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.