| # | id | data_source | prompt | ability | reward_model | extra_info | image |
|---|
1
|
566667
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nProvide a concise explanation of the image contents."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. The large white bird's reflection shows in the water.\n2. A white bird preparing to catch something in the water\n3. A white bird is flying off the water surface.\n4. A large white bird goes across the water\n5. White bird walking across wet sand."
}
|
{
"split": "train",
"index": 66667,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. The large white bird's reflection shows in the water.\n2. A white bird preparing to catch something in the water\n3. A white bird is flying off the water surface.\n4. A large white bird goes across the water\n5. White bird walking across wet sand.",
"question": "<image>\nProvide a concise explanation of the image contents.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
2
|
566668
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nWrite a concise but descriptive caption for the image."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. A middle-aged woman stands at a green podium with the name \"SVA Theatre\" written on it.\n2. A woman, wearing glasses, is standing behind a green podium.\n3. A woman stands at a podium that reads, \"SVA THEATRE.\"\n4. A female speaker at the SVA Theatre is giving a speech.\n5. A woman in speaking in a theater."
}
|
{
"split": "train",
"index": 66668,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. A middle-aged woman stands at a green podium with the name \"SVA Theatre\" written on it.\n2. A woman, wearing glasses, is standing behind a green podium.\n3. A woman stands at a podium that reads, \"SVA THEATRE.\"\n4. A female speaker at the SVA Theatre is giving a speech.\n5. A woman in speaking in a theater.",
"question": "<image>\nWrite a concise but descriptive caption for the image.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
3
|
566669
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nWrite a brief caption suitable for quick understanding."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. There is a woman at the window of a building getting ready to hang a small piece of clothing on a clothesline right outside, which already contains some articles of clothing.\n2. A person in a tan and blue sweater hanging clothes on a clothesline outside the window of her building.\n3. A woman is standing at her window hanging clothes to dry from a clothesline.\n4. Woman in the window just trying to dry some socks.\n5. A woman hangs socks on a line outside her window."
}
|
{
"split": "train",
"index": 66669,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. There is a woman at the window of a building getting ready to hang a small piece of clothing on a clothesline right outside, which already contains some articles of clothing.\n2. A person in a tan and blue sweater hanging clothes on a clothesline outside the window of her building.\n3. A woman is standing at her window hanging clothes to dry from a clothesline.\n4. Woman in the window just trying to dry some socks.\n5. A woman hangs socks on a line outside her window.",
"question": "<image>\nWrite a brief caption suitable for quick understanding.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
4
|
566670
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nExplain what is visible in the image in a concise way."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. A woman in a dark blue running outfit runs through a field of purple flowers.\n2. A woman is running through a field in the rain.\n3. A woman jogs through a field of wild flowers.\n4. Woman running through field in rain.\n5. A woman jogging through a field."
}
|
{
"split": "train",
"index": 66670,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. A woman in a dark blue running outfit runs through a field of purple flowers.\n2. A woman is running through a field in the rain.\n3. A woman jogs through a field of wild flowers.\n4. Woman running through field in rain.\n5. A woman jogging through a field.",
"question": "<image>\nExplain what is visible in the image in a concise way.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
5
|
566671
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nWrite a short caption that explains what is visible."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. A bicycle racer in full helmet speeds through a dim forest.\n2. A mountain biker wearing white races through a turn.\n3. A biker dressed in white rides through the forest.\n4. A person on a motorbike, riding through the woods\n5. Riding a bike through the woods"
}
|
{
"split": "train",
"index": 66671,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. A bicycle racer in full helmet speeds through a dim forest.\n2. A mountain biker wearing white races through a turn.\n3. A biker dressed in white rides through the forest.\n4. A person on a motorbike, riding through the woods\n5. Riding a bike through the woods",
"question": "<image>\nWrite a short caption that explains what is visible.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
6
|
566672
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nDescribe the image in one or two paragraphs."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. A woman wearing blue pants and a blue and pink shirt with a white cloth over her head is kneeling on front of several buckets and containers on the floor.\n2. A person is squatting down near buckets, holding something in one hand.\n3. Lady with a white towel on her head selling at an open air marketplace.\n4. An oriental woman cleaning wood outside from a bucket.\n5. Working women holding a piece of wood."
}
|
{
"split": "train",
"index": 66672,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. A woman wearing blue pants and a blue and pink shirt with a white cloth over her head is kneeling on front of several buckets and containers on the floor.\n2. A person is squatting down near buckets, holding something in one hand.\n3. Lady with a white towel on her head selling at an open air marketplace.\n4. An oriental woman cleaning wood outside from a bucket.\n5. Working women holding a piece of wood.",
"question": "<image>\nDescribe the image in one or two paragraphs.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
7
|
566673
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nSummarize the image clearly without excessive detail."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. A man in a white shirt and brown pants pushes a cart.\n2. A man pushes his walker passed a 'Model Express'.\n3. An old man pushes a cart down the sidewalk.\n4. An old man pushes a cart past a shoe store.\n5. Old man pushing cart in front noodle shop."
}
|
{
"split": "train",
"index": 66673,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. A man in a white shirt and brown pants pushes a cart.\n2. A man pushes his walker passed a 'Model Express'.\n3. An old man pushes a cart down the sidewalk.\n4. An old man pushes a cart past a shoe store.\n5. Old man pushing cart in front noodle shop.",
"question": "<image>\nSummarize the image clearly without excessive detail.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
8
|
566674
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nWrite a concise visual summary of the image."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. A man in a cowboy hat rides a rodeo horse, as another man stands inside the enclosed area and a group of people stand outside the gates to watch.\n2. A man, wearing cowboy attire, is riding a horse in a rodeo, while a crowd behind a barricaded fence surrounds him.\n3. A cowboy rides a bucking bronco in an arena while people in the stands watch him.\n4. A cowboy desperately tries to hang on to a bucking pinto horse during a rodeo.\n5. A cowboy riding a bucking white, black and brown horse."
}
|
{
"split": "train",
"index": 66674,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. A man in a cowboy hat rides a rodeo horse, as another man stands inside the enclosed area and a group of people stand outside the gates to watch.\n2. A man, wearing cowboy attire, is riding a horse in a rodeo, while a crowd behind a barricaded fence surrounds him.\n3. A cowboy rides a bucking bronco in an arena while people in the stands watch him.\n4. A cowboy desperately tries to hang on to a bucking pinto horse during a rodeo.\n5. A cowboy riding a bucking white, black and brown horse.",
"question": "<image>\nWrite a concise visual summary of the image.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
9
|
566675
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nProvide a short summary of the visual scene."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. The man in the black suit is in the water on a white and red board with a sail.\n2. A man windsurfs with a group of other windsurfers.\n3. A man in black swimming gear parasails.\n4. The man is on a ski boat in the water.\n5. A man windsurfing"
}
|
{
"split": "train",
"index": 66675,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. The man in the black suit is in the water on a white and red board with a sail.\n2. A man windsurfs with a group of other windsurfers.\n3. A man in black swimming gear parasails.\n4. The man is on a ski boat in the water.\n5. A man windsurfing",
"question": "<image>\nProvide a short summary of the visual scene.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|
10
|
566676
|
flickr30k
|
{
"role": [
"user"
],
"content": [
"<image>\nWrite a concise but descriptive caption for the image."
]
}
|
captioning_IF
|
{
"style": "rule",
"ground_truth": "Five independent valid human captions:\n1. An escalator with a middle-aged woman, wearing a black shirt, going up and an elderly woman, wearing a red shirt, going down.\n2. A young woman in a black shirt is taking an elevator down with her luggage, and is gazing back in a distressed way.\n3. One woman with a black top and luggage is placing her hand on her neck while traveling up an escalator.\n4. A woman wearing a black t-shirt appears to being going up a escalator.\n5. Various individuals riding escalators."
}
|
{
"split": "train",
"index": 66676,
"domain": "captioning_IF",
"answer": "Five independent valid human captions:\n1. An escalator with a middle-aged woman, wearing a black shirt, going up and an elderly woman, wearing a red shirt, going down.\n2. A young woman in a black shirt is taking an elevator down with her luggage, and is gazing back in a distressed way.\n3. One woman with a black top and luggage is placing her hand on her neck while traveling up an escalator.\n4. A woman wearing a black t-shirt appears to being going up a escalator.\n5. Various individuals riding escalators.",
"question": "<image>\nWrite a concise but descriptive caption for the image.",
"reward_type": "llm_judge",
"tolerance": null
}
|
|