Datasets:

data_source
stringclasses
1 value
prompt
listlengths
1
1
ability
stringclasses
1 value
reward_model
dict
extra_info
dict
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to shelf 1", "style": "rule" }
{ "index": 0, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to sofa 1", "style": "rule" }
{ "index": 1, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take pillow 1 from sofa 1", "style": "rule" }
{ "index": 2, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to armchair 1", "style": "rule" }
{ "index": 3, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move pillow 1 to armchair 1", "style": "rule" }
{ "index": 4, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine armchair 1", "style": "rule" }
{ "index": 5, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to shelf 1", "style": "rule" }
{ "index": 6, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take box 2 from shelf 1", "style": "rule" }
{ "index": 7, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine box 2", "style": "rule" }
{ "index": 8, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine shelf 1", "style": "rule" }
{ "index": 9, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move box 2 to shelf 1", "style": "rule" }
{ "index": 10, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to shelf 2", "style": "rule" }
{ "index": 11, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take statue 3 from shelf 2", "style": "rule" }
{ "index": 12, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to sidetable 1", "style": "rule" }
{ "index": 13, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move statue 3 to sidetable 1", "style": "rule" }
{ "index": 14, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine sidetable 1", "style": "rule" }
{ "index": 15, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to cabinet 1", "style": "rule" }
{ "index": 16, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open cabinet 1", "style": "rule" }
{ "index": 17, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine cabinet 4", "style": "rule" }
{ "index": 18, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open cabinet 4", "style": "rule" }
{ "index": 19, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take soapbottle 3 from cabinet 4", "style": "rule" }
{ "index": 20, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to toilet 1", "style": "rule" }
{ "index": 21, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move soapbottle 3 to toilet 1", "style": "rule" }
{ "index": 22, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine toilet 1", "style": "rule" }
{ "index": 23, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to desk 1", "style": "rule" }
{ "index": 24, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take cd 1 from desk 1", "style": "rule" }
{ "index": 25, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to dresser 1", "style": "rule" }
{ "index": 26, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move cd 1 to dresser 1", "style": "rule" }
{ "index": 27, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine dresser 1", "style": "rule" }
{ "index": 28, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "look", "style": "rule" }
{ "index": 29, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to countertop 1", "style": "rule" }
{ "index": 30, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to countertop 2", "style": "rule" }
{ "index": 31, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to countertop 3", "style": "rule" }
{ "index": 32, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to stoveburner 1", "style": "rule" }
{ "index": 33, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take kettle 2 from stoveburner 1", "style": "rule" }
{ "index": 34, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to shelf 1", "style": "rule" }
{ "index": 35, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move kettle 2 to shelf 1", "style": "rule" }
{ "index": 36, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine shelf 1", "style": "rule" }
{ "index": 37, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to cabinet 1", "style": "rule" }
{ "index": 38, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open cabinet 1", "style": "rule" }
{ "index": 39, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take spraybottle 1 from cabinet 1", "style": "rule" }
{ "index": 40, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to garbagecan 1", "style": "rule" }
{ "index": 41, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move spraybottle 1 to garbagecan 1", "style": "rule" }
{ "index": 42, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine garbagecan 1", "style": "rule" }
{ "index": 43, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "look", "style": "rule" }
{ "index": 44, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to countertop 1", "style": "rule" }
{ "index": 45, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to dresser 1", "style": "rule" }
{ "index": 46, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to diningtable 1", "style": "rule" }
{ "index": 47, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take mug 1 from diningtable 1", "style": "rule" }
{ "index": 48, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to dresser 1", "style": "rule" }
{ "index": 49, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move mug 1 to dresser 1", "style": "rule" }
{ "index": 50, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine dresser 1", "style": "rule" }
{ "index": 51, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 1", "style": "rule" }
{ "index": 52, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 1", "style": "rule" }
{ "index": 53, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 2", "style": "rule" }
{ "index": 54, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 3", "style": "rule" }
{ "index": 55, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 3", "style": "rule" }
{ "index": 56, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take toiletpaper 1 from drawer 3", "style": "rule" }
{ "index": 57, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to toiletpaperhanger 1", "style": "rule" }
{ "index": 58, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move toiletpaper 1 to toiletpaperhanger 1", "style": "rule" }
{ "index": 59, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine toiletpaperhanger 1", "style": "rule" }
{ "index": 60, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 1", "style": "rule" }
{ "index": 61, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 1", "style": "rule" }
{ "index": 62, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 2", "style": "rule" }
{ "index": 63, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 2", "style": "rule" }
{ "index": 64, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 3", "style": "rule" }
{ "index": 65, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 4", "style": "rule" }
{ "index": 66, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine drawer 4", "style": "rule" }
{ "index": 67, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 5", "style": "rule" }
{ "index": 68, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 5", "style": "rule" }
{ "index": 69, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take tissuebox 1 from drawer 5", "style": "rule" }
{ "index": 70, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go\\ to\\ dresser\\ 1", "style": "rule" }
{ "index": 71, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to dresser 1", "style": "rule" }
{ "index": 72, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move tissuebox 1 to dresser 1", "style": "rule" }
{ "index": 73, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine dresser 1", "style": "rule" }
{ "index": 74, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 1", "style": "rule" }
{ "index": 75, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 1", "style": "rule" }
{ "index": 76, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 2", "style": "rule" }
{ "index": 77, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 2", "style": "rule" }
{ "index": 78, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 3", "style": "rule" }
{ "index": 79, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 3", "style": "rule" }
{ "index": 80, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 4", "style": "rule" }
{ "index": 81, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 5", "style": "rule" }
{ "index": 82, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 5", "style": "rule" }
{ "index": 83, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 6", "style": "rule" }
{ "index": 84, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 6", "style": "rule" }
{ "index": 85, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 7", "style": "rule" }
{ "index": 86, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 7", "style": "rule" }
{ "index": 87, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to drawer 8", "style": "rule" }
{ "index": 88, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open drawer 8", "style": "rule" }
{ "index": 89, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to countertop 1", "style": "rule" }
{ "index": 90, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "take spraybottle 2 from countertop 1", "style": "rule" }
{ "index": 91, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to garbagecan 1", "style": "rule" }
{ "index": 92, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "move spraybottle 2 to garbagecan 1", "style": "rule" }
{ "index": 93, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "examine garbagecan 1", "style": "rule" }
{ "index": 94, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to cabinet 1", "style": "rule" }
{ "index": 95, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open cabinet 1", "style": "rule" }
{ "index": 96, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to cabinet 2", "style": "rule" }
{ "index": 97, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "open cabinet 2", "style": "rule" }
{ "index": 98, "split": "train" }
[ { "content": "You are an agent playing a TextWorld kitchen task.\nPlease read the context and your previous actions with environment feedback carefully. Each previous action is separated by [SEP].\nPick EXACTLY ONE next action from the admissible command list.\nReturn only the chosen action string as \\boxed{<a...
math
{ "ground_truth": "go to cabinet 3", "style": "rule" }
{ "index": 99, "split": "train" }