qid string | query string | messages list | videos list | meta dict |
|---|---|---|---|---|
11372_fig | The person cooking on the stove is eating food in a kitchen with white cabinets. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"TIY7F.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "TIY7F",
"query_pos_video_id": "GH3D1",
"pair_id": "11372_fig::TIY7F",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "TIY7F",
"top1_gold_label": "neg"
} |
6082_fig | The person uses a tool, possibly a screwdriver, to fix the doorknob. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"OL2JP.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "OL2JP",
"query_pos_video_id": "0IJ8T",
"pair_id": "6082_fig::OL2JP",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "OL2JP",
"top1_gold_label": "neg"
} |
15142_fig | A person with curly brown hair sits on the floor, bringing food to their mouth from a bowl and eating at a moderate pace. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"HFYNW.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "HFYNW",
"query_pos_video_id": "HCM5Y",
"pair_id": "15142_fig::HFYNW",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "HFYNW",
"top1_gold_label": "neg"
} |
7387_fig | The camera captures the moment of two people laughing in a kitchen, with appliances and cabinets visible in the background. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"14A2O.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "14A2O",
"query_pos_video_id": "KNX2Y",
"pair_id": "7387_fig::14A2O",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "14A2O",
"top1_gold_label": "neg"
} |
12797_fig | A person in a pink outfit is seen fixing their hair. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"0PCAF.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "0PCAF",
"query_pos_video_id": "RYHHP",
"pair_id": "12797_fig::0PCAF",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "0PCAF",
"top1_gold_label": "neg"
} |
13545_fig | A person turns on the light next to a trash can with a white bag. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"BMPEV.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "BMPEV",
"query_pos_video_id": "47I6H",
"pair_id": "13545_fig::BMPEV",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "BMPEV",
"top1_gold_label": "neg"
} |
5101_fig | A person starts dressing by putting on a shirt in a room with a staircase and a wooden cabinet. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"V7LTE.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "V7LTE",
"query_pos_video_id": "7BFDL",
"pair_id": "5101_fig::V7LTE",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "V7LTE",
"top1_gold_label": "neg"
} |
11784_fig | A woman pours milk from a carton into a glass, with a dog watching nearby. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"TU3IG.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "TU3IG",
"query_pos_video_id": "OKL5S",
"pair_id": "11784_fig::TU3IG",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "TU3IG",
"top1_gold_label": "neg"
} |
6113_fig | A person holding an orange towel and dressed in green shirt and plaid shorts, moves down the hallway. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"FZ0LU.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "FZ0LU",
"query_pos_video_id": "Z5AZW",
"pair_id": "6113_fig::FZ0LU",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "FZ0LU",
"top1_gold_label": "neg"
} |
12817_fig | A person in striped shirt and pink shorts is seen fixing a light while standing on a chair. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"EVHAZ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "EVHAZ",
"query_pos_video_id": "ERC0T",
"pair_id": "12817_fig::EVHAZ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "EVHAZ",
"top1_gold_label": "neg"
} |
13688_fig | The person at the coffee table drinks water and eats chips. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"VXF18.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "VXF18",
"query_pos_video_id": "0DBQD",
"pair_id": "13688_fig::VXF18",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "VXF18",
"top1_gold_label": "neg"
} |
6450_fig | A person with a black tank top and pink leggings gets up and opens the door. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"NZ6PU.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "NZ6PU",
"query_pos_video_id": "U9LKA",
"pair_id": "6450_fig::NZ6PU",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "NZ6PU",
"top1_gold_label": "neg"
} |
14321_fig | Searching bag, man locates book and sets it down. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"F4BJJ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "F4BJJ",
"query_pos_video_id": "2LCLG",
"pair_id": "14321_fig::F4BJJ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "F4BJJ",
"top1_gold_label": "neg"
} |
6936_fig | A person wearing glasses and a blue hoodie begins smiling into the mirror as the camera zooms in. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"I6PBQ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "I6PBQ",
"query_pos_video_id": "1YBKW",
"pair_id": "6936_fig::I6PBQ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "I6PBQ",
"top1_gold_label": "neg"
} |
3791_fig | The camera focuses on a person wearing a brown shirt as they close a box on the floor. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"BUBT0.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "BUBT0",
"query_pos_video_id": "T8B7B",
"pair_id": "3791_fig::BUBT0",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "BUBT0",
"top1_gold_label": "neg"
} |
13994_fig | The person, dressed in a black top, opens a box with two tubes sticking out. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"IV7IZ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "IV7IZ",
"query_pos_video_id": "2MAZY",
"pair_id": "13994_fig::IV7IZ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "IV7IZ",
"top1_gold_label": "neg"
} |
15129_fig | A person is seen throwing shoes inside a room with a wooden door. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"O7N6K.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "O7N6K",
"query_pos_video_id": "CH8XJ",
"pair_id": "15129_fig::O7N6K",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "O7N6K",
"top1_gold_label": "neg"
} |
5477_fig | a person opens a door in an interior space with a white couch and pillows. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"1MU5L.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "1MU5L",
"query_pos_video_id": "ME7II",
"pair_id": "5477_fig::1MU5L",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "1MU5L",
"top1_gold_label": "neg"
} |
8427_fig | a person is seen eating a chip near a window with white bars and a curtain. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"1JGRO.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "1JGRO",
"query_pos_video_id": "A42FN",
"pair_id": "8427_fig::1JGRO",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "1JGRO",
"top1_gold_label": "neg"
} |
8647_fig | The person is a man wearing glasses and a pink shirt as he starts eating. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"C5BMA.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "C5BMA",
"query_pos_video_id": "SNUVF",
"pair_id": "8647_fig::C5BMA",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "C5BMA",
"top1_gold_label": "neg"
} |
9614_fig | the camera focuses on a person wearing a pink shirt walking towards a doorway. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"VGTX8.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "VGTX8",
"query_pos_video_id": "S29H6",
"pair_id": "9614_fig::VGTX8",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "VGTX8",
"top1_gold_label": "neg"
} |
7391_fig | The camera focuses on a man sitting cross-legged on the floor with a blue bag visible in the foreground. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"O7NI6.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "O7NI6",
"query_pos_video_id": "IS3OX",
"pair_id": "7391_fig::O7NI6",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "O7NI6",
"top1_gold_label": "neg"
} |
11334_fig | A person puts a black bag on one shoulder on a bed with white sheets. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"TZ27E.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "TZ27E",
"query_pos_video_id": "2OWA8",
"pair_id": "11334_fig::TZ27E",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "TZ27E",
"top1_gold_label": "neg"
} |
12675_fig | A person in a gray shirt pours a glass of water while sitting next to a bed with a laptop on it. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"KX8VW.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "KX8VW",
"query_pos_video_id": "5JVMD",
"pair_id": "12675_fig::KX8VW",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "KX8VW",
"top1_gold_label": "neg"
} |
7323_fig | Engrossed in a game, a person lies on a bed, shifting slightly while making repetitive thumb gestures. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"8CQ9Y.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "8CQ9Y",
"query_pos_video_id": "WWPEN",
"pair_id": "7323_fig::8CQ9Y",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "8CQ9Y",
"top1_gold_label": "neg"
} |
13080_fig | The person rises from the chair in the indoor setting. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"NWFOF.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "NWFOF",
"query_pos_video_id": "4VYE5",
"pair_id": "13080_fig::NWFOF",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "NWFOF",
"top1_gold_label": "neg"
} |
10823_fig | The person starts laughing while reading a book at the table. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"57QQ7.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "57QQ7",
"query_pos_video_id": "M2DD1",
"pair_id": "10823_fig::57QQ7",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "57QQ7",
"top1_gold_label": "neg"
} |
13727_fig | The person begins undressing, starting with the removal of a dark-colored shirt and shorts. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"XTACZ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "XTACZ",
"query_pos_video_id": "BYDSE",
"pair_id": "13727_fig::XTACZ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "XTACZ",
"top1_gold_label": "neg"
} |
7366_fig | Person places dishes in cabinet, maintaining still kitchen scene. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"GAM0E.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "GAM0E",
"query_pos_video_id": "8PXPO",
"pair_id": "7366_fig::GAM0E",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "GAM0E",
"top1_gold_label": "neg"
} |
15809_fig | A person, lying in bed, stretches, yawns, and gets up. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"99596.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "99596",
"query_pos_video_id": "XDT45",
"pair_id": "15809_fig::99596",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "99596",
"top1_gold_label": "neg"
} |
15406_fig | A person is seen in the bathroom, drinking from a glass with a mirror reflecting the scene above the sink. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"NEFHO.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "NEFHO",
"query_pos_video_id": "F76GD",
"pair_id": "15406_fig::NEFHO",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "NEFHO",
"top1_gold_label": "neg"
} |
13995_fig | The woman, wearing a black shirt, opens a white box and retrieves two small black objects. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"8HHI3.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "8HHI3",
"query_pos_video_id": "2MAZY",
"pair_id": "13995_fig::8HHI3",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "8HHI3",
"top1_gold_label": "neg"
} |
8191_fig | Person pulls open door, revealing bright light, and stands in doorway with phone. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"3FMAS.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "3FMAS",
"query_pos_video_id": "PB7PQ",
"pair_id": "8191_fig::3FMAS",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "3FMAS",
"top1_gold_label": "neg"
} |
15945_fig | Person, with closed laptop on lap, opens it using right hand. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"0R0DJ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "0R0DJ",
"query_pos_video_id": "JBJGX",
"pair_id": "15945_fig::0R0DJ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "0R0DJ",
"top1_gold_label": "neg"
} |
7851_fig | A man in a blue shirt sits on a red chair and opens his laptop. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"YDTH0.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "YDTH0",
"query_pos_video_id": "0W4F6",
"pair_id": "7851_fig::YDTH0",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "YDTH0",
"top1_gold_label": "neg"
} |
5008_fig | The camera focuses on a person working intently on his laptop, dressed in a dark-colored shirt. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"PWR6D.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "PWR6D",
"query_pos_video_id": "MY5CQ",
"pair_id": "5008_fig::PWR6D",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "PWR6D",
"top1_gold_label": "neg"
} |
11812_fig | The person, dressed in a white hoodie, puts the phone on the table. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"8AKGE.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "8AKGE",
"query_pos_video_id": "11TTU",
"pair_id": "11812_fig::8AKGE",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "8AKGE",
"top1_gold_label": "neg"
} |
5403_fig | A woman in a green top is shown putting her feet up on a box. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"DP7DG.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "DP7DG",
"query_pos_video_id": "DHFA6",
"pair_id": "5403_fig::DP7DG",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "DP7DG",
"top1_gold_label": "neg"
} |
9424_fig | The camera focuses on a person sneezing into a phone in a basement filled with some boxes. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"GZQ0V.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "GZQ0V",
"query_pos_video_id": "TDHM0",
"pair_id": "9424_fig::GZQ0V",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "GZQ0V",
"top1_gold_label": "neg"
} |
12612_fig | A person, wearing glasses, puts on their shoes in a living room. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"ZOTR8.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "ZOTR8",
"query_pos_video_id": "AWGMW",
"pair_id": "12612_fig::ZOTR8",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "ZOTR8",
"top1_gold_label": "neg"
} |
4205_fig | Wrapped person consumes soup, tilting head and bringing spoon to mouth, amidst kitchen counter's clutter. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"HFYNW.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "HFYNW",
"query_pos_video_id": "A1NKP",
"pair_id": "4205_fig::HFYNW",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "HFYNW",
"top1_gold_label": "neg"
} |
10477_fig | The person opening the door is in a room with a clock visible on the wall. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"RVWK8.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "RVWK8",
"query_pos_video_id": "8BBX0",
"pair_id": "10477_fig::RVWK8",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "RVWK8",
"top1_gold_label": "neg"
} |
15490_fig | A man in a dark-colored jacket and light-colored pants, holding a broom, is seen sneezing in an indoor setting. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"3NOML.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "3NOML",
"query_pos_video_id": "C3JAS",
"pair_id": "15490_fig::3NOML",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "3NOML",
"top1_gold_label": "neg"
} |
12543_fig | The camera captures a person eating in a cozy room, with a bed and a desk visible in the background. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"DWBS3.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "DWBS3",
"query_pos_video_id": "HUNW6",
"pair_id": "12543_fig::DWBS3",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "DWBS3",
"top1_gold_label": "neg"
} |
9086_fig | A person takes a picture with their cellphone camera, wearing a black outfit and standing in a kitchen with two windows covered in curtains in the background. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"7SUTI.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "7SUTI",
"query_pos_video_id": "X8XQE",
"pair_id": "9086_fig::7SUTI",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "7SUTI",
"top1_gold_label": "neg"
} |
8246_fig | a person is sitting on a pillow on a chair, with a laptop open and turned on in front of them. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"AJRIJ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "AJRIJ",
"query_pos_video_id": "IG1YT",
"pair_id": "8246_fig::AJRIJ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "AJRIJ",
"top1_gold_label": "neg"
} |
6181_fig | A person, wearing glasses and holding a white towel close to their face, remains mostly still with slight head and hand movements in a room with wood-paneled walls. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"901GH.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "901GH",
"query_pos_video_id": "XJ4T3",
"pair_id": "6181_fig::901GH",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "901GH",
"top1_gold_label": "neg"
} |
13935_fig | A person is seen taking medicine with a door visible in the background. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"857LD.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "857LD",
"query_pos_video_id": "NBE4C",
"pair_id": "13935_fig::857LD",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "857LD",
"top1_gold_label": "neg"
} |
6144_fig | A person in a white tank top and black pants opens up a plastic bag. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"O76N2.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "O76N2",
"query_pos_video_id": "XTC0F",
"pair_id": "6144_fig::O76N2",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "O76N2",
"top1_gold_label": "neg"
} |
12909_fig | The camera focuses on a person running down the stairs, dressed in a red shirt and blue shorts. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"RZLAZ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "RZLAZ",
"query_pos_video_id": "RKIC6",
"pair_id": "12909_fig::RZLAZ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "RZLAZ",
"top1_gold_label": "neg"
} |
10523_fig | The person, holding a pink blanket or cloth, sits on the chair. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"ZAJAJ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "ZAJAJ",
"query_pos_video_id": "8Z9GW",
"pair_id": "10523_fig::ZAJAJ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "ZAJAJ",
"top1_gold_label": "neg"
} |
14716_fig | the camera focuses on a young man sitting in the chair, dressed in a gray t-shirt and black shorts, with a laptop. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"6LUM5.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "6LUM5",
"query_pos_video_id": "LBG7L",
"pair_id": "14716_fig::6LUM5",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "6LUM5",
"top1_gold_label": "neg"
} |
9414_fig | a person, who is a young woman, is undressing and holding a red object. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"XKLW2.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "XKLW2",
"query_pos_video_id": "2NZ5C",
"pair_id": "9414_fig::XKLW2",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "XKLW2",
"top1_gold_label": "neg"
} |
10444_fig | A woman in a white long-sleeved shirt and blue jeans is opening the door. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"YM213.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "YM213",
"query_pos_video_id": "AJZYE",
"pair_id": "10444_fig::YM213",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "YM213",
"top1_gold_label": "neg"
} |
5506_fig | The person closes the door in a room with a bed as the main feature. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"VWN13.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "VWN13",
"query_pos_video_id": "R5L98",
"pair_id": "5506_fig::VWN13",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "VWN13",
"top1_gold_label": "neg"
} |
15399_fig | A person in light clothing enters the hallway, running towards an opening door. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"Z5FNI.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "Z5FNI",
"query_pos_video_id": "2FL0X",
"pair_id": "15399_fig::Z5FNI",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "Z5FNI",
"top1_gold_label": "neg"
} |
6874_fig | A person puts clothes in a box among bottles labeled 'Green Works'. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"OEE36.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "OEE36",
"query_pos_video_id": "A58GR",
"pair_id": "6874_fig::OEE36",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "OEE36",
"top1_gold_label": "neg"
} |
16016_fig | One person, a young boy, is eating in an indoor setting, with a doorway leading to another room. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"8M0ZV.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "8M0ZV",
"query_pos_video_id": "YHPLU",
"pair_id": "16016_fig::8M0ZV",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "8M0ZV",
"top1_gold_label": "neg"
} |
5585_fig | a person is eating a sandwich while sitting on a couch in a living room with a kitchen area in the background. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"B7DNX.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "B7DNX",
"query_pos_video_id": "ZCH7N",
"pair_id": "5585_fig::B7DNX",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "B7DNX",
"top1_gold_label": "neg"
} |
9881_fig | A person in a blue shirt and plaid pants awakens on the floor. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"PW6CO.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "PW6CO",
"query_pos_video_id": "4N06O",
"pair_id": "9881_fig::PW6CO",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "PW6CO",
"top1_gold_label": "neg"
} |
8094_fig | The person is shown in motion as they undress, possibly walking or moving their arms. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"Q7RK6.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "Q7RK6",
"query_pos_video_id": "O2294",
"pair_id": "8094_fig::Q7RK6",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "Q7RK6",
"top1_gold_label": "neg"
} |
10748_fig | a person, who is the main focus, puts clothes into a washer. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"APJQ8.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "APJQ8",
"query_pos_video_id": "APTS7",
"pair_id": "10748_fig::APJQ8",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "APJQ8",
"top1_gold_label": "neg"
} |
9636_fig | Person in kitchen attire cooks, wielding a spoon over a heating pot. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"DUEEE.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "DUEEE",
"query_pos_video_id": "IBNKL",
"pair_id": "9636_fig::DUEEE",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "DUEEE",
"top1_gold_label": "neg"
} |
8356_fig | A man in a black hoodie takes out a baggy from the refrigerator. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"NH9JS.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "NH9JS",
"query_pos_video_id": "HZTB1",
"pair_id": "8356_fig::NH9JS",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "NH9JS",
"top1_gold_label": "neg"
} |
6050_fig | A person was laughing at something on television in a room with a pinkish hue. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"3D81G.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "3D81G",
"query_pos_video_id": "13IOT",
"pair_id": "6050_fig::3D81G",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "3D81G",
"top1_gold_label": "neg"
} |
7669_fig | A person is sneezing while wearing a blue shirt. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"1TAMK.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "1TAMK",
"query_pos_video_id": "X6JKA",
"pair_id": "7669_fig::1TAMK",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "1TAMK",
"top1_gold_label": "neg"
} |
9717_fig | A person with purple hair, in a laundry room, remains crouched and laughs with an open mouth and squinted eyes. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"J6P6H.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "J6P6H",
"query_pos_video_id": "2S6V7",
"pair_id": "9717_fig::J6P6H",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "J6P6H",
"top1_gold_label": "neg"
} |
14501_fig | A person is sitting at a table with a blue tablecloth, closing a laptop. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"H69BE.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "H69BE",
"query_pos_video_id": "P2HZG",
"pair_id": "14501_fig::H69BE",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "H69BE",
"top1_gold_label": "neg"
} |
8271_fig | A man sets down and readjusts his laptop on a table. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"HL4V4.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "HL4V4",
"query_pos_video_id": "M2XIS",
"pair_id": "8271_fig::HL4V4",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "HL4V4",
"top1_gold_label": "neg"
} |
12443_fig | A person in orange enters a sunlit doorway, looking down at a phone, then pauses to gaze out. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"3FMAS.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "3FMAS",
"query_pos_video_id": "C8WLX",
"pair_id": "12443_fig::3FMAS",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "3FMAS",
"top1_gold_label": "neg"
} |
9702_fig | The man opens the blinds and gazes out the window. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"F1C8I.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "F1C8I",
"query_pos_video_id": "S72J9",
"pair_id": "9702_fig::F1C8I",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "F1C8I",
"top1_gold_label": "neg"
} |
11086_fig | A person puts a blanket to their legs, dressed in a plaid shirt and shorts, in a cluttered room. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"0AGCS.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "0AGCS",
"query_pos_video_id": "TIGIP",
"pair_id": "11086_fig::0AGCS",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "0AGCS",
"top1_gold_label": "neg"
} |
6278_fig | a person is fixing some string lights in a room with a couch and a window. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"0CX32.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "0CX32",
"query_pos_video_id": "TXJK7",
"pair_id": "6278_fig::0CX32",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "0CX32",
"top1_gold_label": "neg"
} |
9121_fig | A person takes a bag of chips out, revealing a shelf with framed pictures in the background. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"FZIW9.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "FZIW9",
"query_pos_video_id": "583F7",
"pair_id": "9121_fig::FZIW9",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "FZIW9",
"top1_gold_label": "neg"
} |
4548_fig | Young woman in living room, holding a cup, gazes right; no light change observed. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"X0CKA.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "X0CKA",
"query_pos_video_id": "618HP",
"pair_id": "4548_fig::X0CKA",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "X0CKA",
"top1_gold_label": "neg"
} |
10055_fig | A person is washing a window in an entryway with a wood door having a glass panel. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"WESPT.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "WESPT",
"query_pos_video_id": "V149B",
"pair_id": "10055_fig::WESPT",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "WESPT",
"top1_gold_label": "neg"
} |
11039_fig | A person consumes snacks from a bowl as another person prepares nearby. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"IO3CI.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "IO3CI",
"query_pos_video_id": "TK76G",
"pair_id": "11039_fig::IO3CI",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "IO3CI",
"top1_gold_label": "neg"
} |
5226_fig | A person, holding a pair of shoes, enters the frame and puts them away. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"A2EY4.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "A2EY4",
"query_pos_video_id": "BQAUC",
"pair_id": "5226_fig::A2EY4",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "A2EY4",
"top1_gold_label": "neg"
} |
11287_fig | A person in a rush starts running through an interior space, the background showing a white wall and a bedroom with a bed having a blue blanket. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"DM2HW.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "DM2HW",
"query_pos_video_id": "PWC0F",
"pair_id": "11287_fig::DM2HW",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "DM2HW",
"top1_gold_label": "neg"
} |
15925_fig | A person, bald and middle-aged, wearing a yellow and white striped shirt, sits on a red chair while reaching out to put the sandwich back on the ledge. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"SNUVF.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "SNUVF",
"query_pos_video_id": "Q6IU9",
"pair_id": "15925_fig::SNUVF",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "SNUVF",
"top1_gold_label": "neg"
} |
10632_fig | A person, wearing a dark blue shirt and khakis, approaches a closed door, covers their mouth as they sneeze, and exits the frame. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"G7LU9.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "G7LU9",
"query_pos_video_id": "EBTD6",
"pair_id": "10632_fig::G7LU9",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "G7LU9",
"top1_gold_label": "neg"
} |
6435_fig | A person takes off their shoes and sits on a bed with a red floor. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"13IJA.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "13IJA",
"query_pos_video_id": "NV4FC",
"pair_id": "6435_fig::13IJA",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "13IJA",
"top1_gold_label": "neg"
} |
5949_fig | A man in a black cap tidies up the table. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"ICALV.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "ICALV",
"query_pos_video_id": "FM078",
"pair_id": "5949_fig::ICALV",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "ICALV",
"top1_gold_label": "neg"
} |
5549_fig | A man in a plaid shirt opens a cabinet. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"16D5O.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "16D5O",
"query_pos_video_id": "9335E",
"pair_id": "5549_fig::16D5O",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "16D5O",
"top1_gold_label": "neg"
} |
11816_fig | A young woman, initially adjusting her hair, gradually smiles and gestures with excitement in a bathroom mirror. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"0PCAF.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "0PCAF",
"query_pos_video_id": "DSWWV",
"pair_id": "11816_fig::0PCAF",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "0PCAF",
"top1_gold_label": "neg"
} |
8626_fig | Person prepares and consumes a sandwich on camera. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"VDUO1.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "VDUO1",
"query_pos_video_id": "VLYM4",
"pair_id": "8626_fig::VDUO1",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "VDUO1",
"top1_gold_label": "neg"
} |
12629_fig | The person sits down on a chair in front of a desk with a laptop on it. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"PKRP5.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "PKRP5",
"query_pos_video_id": "DSGVL",
"pair_id": "12629_fig::PKRP5",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "PKRP5",
"top1_gold_label": "neg"
} |
5317_fig | a person stands in a living room, holding a pink mirror, and gazes at their reflection. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"OLSMC.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "OLSMC",
"query_pos_video_id": "49QXY",
"pair_id": "5317_fig::OLSMC",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "OLSMC",
"top1_gold_label": "neg"
} |
8720_fig | A person pours water into a white cup. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"HOLED.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "HOLED",
"query_pos_video_id": "FP7N3",
"pair_id": "8720_fig::HOLED",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "HOLED",
"top1_gold_label": "neg"
} |
8456_fig | a person is drinking a glass of water while wearing a striped shirt. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"6YGBD.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "6YGBD",
"query_pos_video_id": "PZ2W1",
"pair_id": "8456_fig::6YGBD",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "6YGBD",
"top1_gold_label": "neg"
} |
15595_fig | The camera captures the moment a person closes the door in this cozy interior space. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"0WLCJ.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "0WLCJ",
"query_pos_video_id": "J8781",
"pair_id": "15595_fig::0WLCJ",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "0WLCJ",
"top1_gold_label": "neg"
} |
8630_fig | A person is eating a sandwich while sitting on a chair. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"QH3FK.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "QH3FK",
"query_pos_video_id": "E4C1C",
"pair_id": "8630_fig::QH3FK",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "QH3FK",
"top1_gold_label": "neg"
} |
5389_fig | Person, holding a pillow and laughing, moves towards camera as door swings open. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"FBIVN.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "FBIVN",
"query_pos_video_id": "SMVKB",
"pair_id": "5389_fig::FBIVN",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "FBIVN",
"top1_gold_label": "neg"
} |
14422_fig | A man in gray shirt and black pants, wearing glasses, stands in front of a bathroom scene and closes the door behind him. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"YKT8I.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "YKT8I",
"query_pos_video_id": "9JHH9",
"pair_id": "14422_fig::YKT8I",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "YKT8I",
"top1_gold_label": "neg"
} |
13210_fig | one person, dressed in a black top, runs into the entryway with a towel. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"2074D.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "2074D",
"query_pos_video_id": "PROQY",
"pair_id": "13210_fig::2074D",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "2074D",
"top1_gold_label": "neg"
} |
9716_fig | A man in a black shirt starts laughing in a room with a mirror on the wall. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"JSLW5.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "JSLW5",
"query_pos_video_id": "64IAU",
"pair_id": "9716_fig::JSLW5",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "JSLW5",
"top1_gold_label": "neg"
} |
5202_fig | a person walks through a doorway eating food, wearing a blue and white striped shirt. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"QXEWB.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "QXEWB",
"query_pos_video_id": "QO6RV",
"pair_id": "5202_fig::QXEWB",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "QXEWB",
"top1_gold_label": "neg"
} |
10412_fig | The person turning on a light is a man dressed in a white shirt and gray pants. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"W7WJR.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "W7WJR",
"query_pos_video_id": "5LETG",
"pair_id": "10412_fig::W7WJR",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "W7WJR",
"top1_gold_label": "neg"
} |
15546_fig | A person is seen eating a sandwich in front of a doorway. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"1H6PS.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "1H6PS",
"query_pos_video_id": "42MC3",
"pair_id": "15546_fig::1H6PS",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "1H6PS",
"top1_gold_label": "neg"
} |
8552_fig | A person opens the door to the refrigerator, revealing its contents, while wearing a striped shirt and a skirt. | [
{
"role": "system",
"content": "You are a video retrieval assistant. Your task is to analyze a retrieved video against the user query. Inside <think>...</think>, perform a step by step comparison between the query requirements and the visible evidence in the video. Identify whether a scene corresponding to ... | [
"M0KKW.npy"
] | {
"source": "pairs_with_reasoning",
"gold_label": "neg",
"video_id": "M0KKW",
"query_pos_video_id": "Q4UO4",
"pair_id": "8552_fig::M0KKW",
"episode_type": "T1",
"num_negs": 1,
"top1_video_id": "M0KKW",
"top1_gold_label": "neg"
} |
End of preview. Expand in Data Studio
VideoSearch-R1 Charades-STA
This repository contains the prepared Charades-STA artifacts used by VideoSearch-R1.
Stage 1 Cold Start SFT
Charades_STA_Stage1_ColdStart.jsonl is the Stage 1 cold-start SFT dataset used to train the VideoSearch-R1 verifier/reasoner on Charades-STA. Each row keeps the ShareGPT-style messages format:
system: task instruction and output schemauser: query plus retrieved video placeholderassistant: reasoning trace and final match decision
The original training-compatible JSON array is also retained at sft_data/train_oneturn.json. Video paths in each row refer to the prepared training video features under train/video_npy_with_meta.
- Downloads last month
- 77