Q2E: Query-to-Event Decomposition for Zero-Shot Multilingual
Collection
Datasets used in the paper: Q2E: Query-to-Event Decomposition for Zero-Shot Multilingual Text-to-Video Retrieval • 37 items • Updated
query stringclasses 259
values | prequel listlengths 0 270 | sequel listlengths 0 162 | during listlengths 0 162 | metadata dict | num_of_frames int64 16 16 | frame_captions listlengths 16 16 | video_id stringlengths 6 45 | frame2video_caption stringlengths 54 2.35k |
|---|---|---|---|---|---|---|---|---|
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=t8lY-rjy7-Y",
"video_description": "9월 12일, 경주에서는 역대 최대 규모인 5.8 지진이 일어났다. 지진으로 서울과 수도권까지 전국이 흔들렸고, 이로 인해 전화와 카카오톡도 모두 먹통이었다. 정부는 지진 발생 9분이 지나서야 긴급재난문자를 발송했고 이마저도 서울과 수도권 지역 국민들은 문자를 받지 못했다. #경주 #지진 #재난문자",
"video_id": "t8lY-rjy7-Y"
} | 16 | [
"The image shows a group of people, likely students, rushing down a hallway and up a staircase. They appear to be in a hurry, possibly evacuating the building. The timestamp on the image reads \"7시 44분,\" which indicates the time as 7:44 AM. The setting seems to be a school or similar institution.",
"The image sh... | t8lY-rjy7-Y | A group of students rush down a hallway and up a staircase to evacuate a building in response to a reported earthquake, with a train station scene outside displaying an earthquake detection and the National Assembly reacting to the situation with a press conference. The scene shifts to a warehouse with a person discussing the impact of the earthquake, followed by a bus driving through debris, and a domestic scene showing a person playing with a dog, indicating a domestic reaction to the earthquake. A governmental press conference follows, discussing the situation and the impact on national security services. A screenshot of text messages on a mobile phone further emphasizes the focus on national security services and the operational status of government institutions. The scene shifts to a person addressing the media in a formal setting, apologizing for an incident involving a person, and finally to a chaotic scene in a school hallway where students are reacting to an unexpected event, with students in a classroom displaying confusion and distress. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=CXe6_CDrdB8",
"video_description": "찐♡기상청사람들 예고 없이 찾아오는 지진! 한 번 발생하면 큰 피해를 주는 지진이기에 감시하는 일은 한시도 소홀히 할 수 없습니다. 그 최전방에서... 1년 365일 24시간 지진화산 감시 및 분석 업무에 최선을 다해 신속·정확하게 알리고자 노력하는 사람이 있습니다. 지진화산 감시 및 분석 업무 9년 차... 반복되는 주간 근무와 야간 근무가 힘들긴 하지만, 누구보다 묵묵하게 자리를 지키고 있는 진득한 뚝배기 같은 지진업무의 대들보!! 지진화산 감시 업무를 담당하는 기상청 사람을 만나고 왔습니다.(^_^)v 우리나라에서 가장 강력했던 규모 5.8의 경주 지진 때 지진감시 및 분석을 수행했던 기상청 사람은 누구? 지진감시 업무를 들어올 때마다 긴장되는 기상청 사람은 누구? 국민의 안전을 누구보다 걱정하는 기상청 사람은 누구? 지진 올바로 알고 지진 발생 시 빠르게 대처하길 바라는 기상청 사람은 누구? 최근 지진 안전을 홍보해 주시는 지진희 배우님을 그 누구보다 고마워하는 기상청 사람은 누구? 만나볼까요 :) #기상청사람들 #지진화산 #행동하는파수꾼",
"video_id": "CXe6_CDrdB8"
} | 16 | [
"In this frame, a man is standing in front of a wall of multiple monitors displaying various data and graphs. He is wearing a blue shirt and a white watch, and appears to be presenting or discussing the content on the screens. The text on the screen reads \"Q 엄무 소개?\" which translates to \"Q Introduction to work?\"... | CXe6_CDrdB8 | A major earthquake occurred in Gyeongju, South Korea with a magnitude of 5.8. The event was the largest to have occurred in Korea, and it was detected through multiple monitoring stations. The earthquake was a significant event for the region, and data analysis was crucial in determining its location and intensity. The data analysis involved experts in monitoring stations who interpreted the seismic waves to provide timely warnings to the public. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=uF3x63vtddc",
"video_description": "\"경주 지진, 지하수는 1~2일 전에 미리 알았다\" 지난 12일과 19일 경주에서 규모 5.8을 비롯한 대형 지진이 잇따라 발생하기 하루 이틀 전 주변 지하수 수위가 크게 상승해, 향후 지진예측에 활용할 수 있다는 주장이 나왔습니다. 정상용 부경대 지구환경과학과 교수는 진앙에서 가장 가까운 경북 경주시 산내면에 있는 지하수 관측소 자료를 분석한 결과, 이 같이 밝혔습니다. 자료를 보면 지난 11일 암반 지하수 수위는 하루 사이에 무려 49㎝ 상승했고, 17일에도 39㎝나 올라갔습니다. 정 교수는 지진에 앞서 암석에 강한 압력이 가해져 지하수 수위가 많이 올라간 것으로 보인다며, 규모 4.5 이상의 지진에만 영향을 받았다고 설명했습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "uF3x63vtddc"
} | 16 | [
"In the image, two news anchors are standing in a news studio. The male anchor is on the left and the female anchor is on the right. They appear to be delivering a news report. The studio background shows multiple screens displaying various images, likely from other news stories or related footage. \n\nThe text on ... | uF3x63vtddc | In the given video, it appears to be documenting the story of an earthquake in Gyeongju, South Korea, which occurred on 1-2 days before the time of 09:34. The footage focuses on a news report that highlighted the early warning signs of the earthquake, including the indication of groundwater being affected. The incident was reportedly related to a machine in Gyeongju that was operational for 1-2 days before the event. The footage shows a series of images and descriptions of the aftermath of the earthquake, including damage to buildings, debris, and flooding. The narrative also includes news segments and footage from news channels, providing context to the incident. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=nln3QFJggKs",
"video_description": "106억 → 58억…경주 지진 피해금액 반토막 잇단 강진으로 100억원이 넘는 것으로 추정됐던 경주시의 공식 피해규모가 58억원 수준에 그치는 것으로 집계됐습니다. 경주시가 지진 발생 뒤 전수조사를 통해 발표한 잠정 추산 피해액은 약 107억원이었지만 사유재산 피해는 대부분 공식 피해액에서 빠졌습니다. 규정상 기와가 떨어지거나 벽체에 금이 가고 담장이 파손된 것은 재산피해 산정되지 않습니다. 이에 따라 사유재산 중에는 지진으로 주택이 완전히 부서지거나 반쯤 허물어진 것만 공식 피해액에 들어갔습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "nln3QFJggKs"
} | 16 | [
"The image shows a news broadcast from Yonhap News TV, featuring two news anchors. The male anchor on the left is dressed in a dark suit and red tie, while the female anchor on the right wears a blue blazer over a pink top. Both are standing in front of a backdrop depicting a nighttime street scene. There is a larg... | nln3QFJggKs | The video is a series of images depicting the aftermath of a massive earthquake in Gyeongju, South Korea, where a 4.5 aftershock has led to significant reductions in damage estimates. The video focuses on:
- A 6-frame series of images showing the aftermath of the earthquake, highlighting the extent of damage to roads, buildings, and infrastructure.
- The damage includes cracked or destroyed buildings, roads, and debris.
- The estimated damage amounts have been reduced from 106 billion to 58 billion won.
- The scene emphasizes the community's efforts to recover and rebuild, with people gathering for aid and assistance.
- The video captures the ongoing process of inspection and repair, highlighting the urgency of addressing the earthquake's impact.
The scenes are set in a predominantly somber and quiet atmosphere, emphasizing the extent of the destruction and the need for immediate attention to prevent further deterioration. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=n6UmQ9Kw2GQ",
"video_description": "경주 지진 이후 여진 307회 발생…\"강도 약해져\" 지난 12일 저녁 발생한 규모 5.8의 경주 지진 탓에 여진이 307차례 발생한 것으로 집계됐습니다. 기상청에 따르면 오늘 오후 4시까지 경주 지진 이후 여진은 규모 1.5∼3.0도 292회 등 모두 307차례 발생했습니다. 기상청 관계자는 \"경주 지진 탓에 여진이 계속 발생하고 있지만 강도가 약해지고 있다\"며 \"앞으로 일어날 여진 규모는 3.0 이하로 예상되기 때문에 그렇게 우려하지 않아도 될 것 같다\"고 말했습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "n6UmQ9Kw2GQ"
} | 16 | [
"The image shows a news broadcast with two news anchors wearing traditional Korean clothing. The screen behind them displays a weather map highlighting a region in South Korea, indicating a severe weather event. The text on the screen reads, \"경주 지진 이후 여진 307회 발생... '강도 약해져'\" which translates to \"Following the Gy... | n6UmQ9Kw2GQ | The video depicts a severe weather event and aftershocks in South Korea, specifically following the Gyeongju earthquake. The narrative emphasizes the aftermath of the earthquake and its ongoing impact on the local community, particularly women. The video focuses on various aspects of their lives, including their daily routines, experiences, and perceptions of the event. The narrative highlights the reduction in intensity of the earthquake over time, suggesting that the community has adapted to the ongoing challenges. The video also mentions the presence of various household items and logistics services, such as delivery times and household essentials. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=uoagwCVoHGM",
"video_description": "경주 지진으로 8명 부상…재산피해 253건 어제 저녁 경주 일대에서 발생한 지진으로 인해 8명이 다치고 카카오톡과 국민안전처 홈페이지 등 통신망이 일시적으로 마비되는 등의 장애가 있었습니다. 국민안전처는 경주 일대에서 오늘과 어제에 걸쳐 일어난 지진으로 인해 80살 여성 황 모 씨 등 8명이 경상을 입고 253건의 재산피해가 신고됐다고 밝혔습니다. 이번 지진으로 인해 KTX 대전~부산 구간이 5분가량 정차하기도 했습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "uoagwCVoHGM"
} | 16 | [
"A news anchor in a purple blazer and light pink top is presenting a story. Behind her is a black-and-white surveillance image showing a scene with water, possibly from a flood, as it flows over a barrier. The image features several potted plants and a railing on the left side. The text on the screen, in Korean, me... | uoagwCVoHGM | The video depicts the aftermath of a significant earthquake in Gyeongju, South Korea, resulting in 8 injuries and 253 property damages. The event occurred after a 6.8 magnitude earthquake hit the city, causing widespread destruction and disruption. The visual details show various scenes of damaged buildings, debris, and injured people. The footage includes footage of emergency responders, cleanup efforts, and a bustling airport scene. The overall context suggests that the earthquake has had a severe impact on the city, including structural damage to buildings and significant loss of life. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=5p1N4MtudiI",
"video_description": "경주 지진 316회 이어가다 현재는 '잠잠' 지난 12일 발생한 규모 5.8의 경주 지진의 여진이 316회 발생한 것으로 집계됐습니다. 기상청은 오늘 오전 11시 기준 여진 횟수가 316회라고 발표했습니다. 규모별 횟수는 1.5에서 3.0이 301차례, 3.0에서 4.0은 14차례, 4.0에서 5.0은 1차례 감지됐습니다. 사흘째 계속된 여진은 일단 멈추는 징후를 보이고 있는 것으로 기상청은 파악하고 있습니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "5p1N4MtudiI"
} | 16 | [
"The image is a still from a news broadcast on EBS TV. Two presenters are standing in front of a cityscape backdrop, featuring a bridge over a river with buildings in the background. The presenters are dressed in traditional Korean hanbok attire. The text at the bottom of the screen in Korean mentions \"The 316th I... | 5p1N4MtudiI | The 316th Invention Exhibition, a notable event in Korea, has been declared "Deadlock" after 31 years without any notable event. The event features a backdrop of traditional Korean attire and is centered on the 316th Invention Exhibition, a showcase of Korean inventions. The exhibition has been marked by severe weather patterns and ongoing seismic activity.
The event is currently surrounded by a news broadcast that highlights the status as "Deadlock," indicating that the weather has worsened and seismic activity continues to rise. Seismic data, including maps, radar images, and radar waveforms, are presented, suggesting that the situation remains critical. The event continues to attract significant public attention with the number 89 being displayed prominently on the screen, possibly representing an ongoing analysis of seismic events. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=roaxxb8zRLU",
"video_description": "국제신문 뉴스&이슈 구독하기 http://goo.gl/FFQL4 국제신문 근교산&기획특집 구독하기 http://goo.gl/SM0fM 국제신문 기사보기 http://iii.im/pk3Z 국제신문 영상팀이 경주 지진피해 복구 현장을 드론으로 담았다.",
"video_id": "roaxxb8zRLU"
} | 16 | [
"The image shows an aerial view of a traditional Korean village with numerous houses featuring black-tiled roofs. The text on the image is in Korean and reads \"하늘에서 본 경주\" (Jeju seen from the sky) and \"9.24일 도요미 경주시 황남동 일대\" (On September 24, Toyomi, Hwangnam-dong area of Gyeongju City). The view captures the den... | roaxxb8zRLU | The video depicts a traditional Korean village, known as Gyeongju, with numerous black-tiled roofs, amidst the scenic landscape of the Korean countryside. The village is the focus of the video, showcasing various scenes of daily life, architectural structures, and maintenance work. The restoration of historical buildings and homes in the village is a significant aspect of the video, highlighting the efforts of the community to preserve cultural heritage. The video transitions from a serene and unobstructed view of the village to a focus on restoration work, showcasing the community's dedication to preserving the region's history and cultural significance.
The video also highlights various aspects of traditional Korean culture, including traditional architecture, festivals, and community activities. The inclusion of natural elements, such as greenery and mountains, emphasizes the natural beauty of the region. The video provides a visual and informational overview of Gyeongju, showcasing its unique cultural heritage and significance.
Throughout the video, the narrative is delivered through the descriptions of individual frames, each focusing on specific aspects of the village, such as architecture, maintenance work, and community activities. The videos transition smoothly, seamlessly moving between these different perspectives, providing a comprehensive and engaging visual narrative of the village and its significance. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=3ojHTwrvx90",
"video_description": "박 대통령, 경주 지진현장 방문 \"특별재난지역 선포\" 박근혜 대통령은 최근 강진이 잇따라 발생한 경북 경주 지역을 찾아 특별재난지역 선포를 적극 검토하라고 지시했습니다. 이어 지진 등 재난 안전 시스템을 원점에서 재점검할 것도 주문했습니다. 박 대통령은 아울러 인근 월성 원자력발전소를 찾아 \"원전은 단 한치의 실수도 용납되지 않는다\"며 지진 방재대책을 철저히 재점검해달라고 당부했습니다. 박 대통령의 지시에 국민안전처 또한 경주시를 특별재난지역으로 선포할 예정입니다. 특별재난지역으로 지정되면 지자체의 피해 복구비 부담이 완화되고, 피해 주민들은 공공요금 감면혜택을 받게됩니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "3ojHTwrvx90"
} | 16 | [
"The current frame features two news anchors presenting a news story. In the background, a group of people is gathered, with one woman in a yellow jacket speaking to the crowd. The text on the screen indicates the news is about President Park visiting a local community, mentioning a special voting area.",
"In the... | 3ojHTwrvx90 | President Park's delegation, consisting of officials and staff in yellow jackets, is actively involved in the voting area for a special election in Jeonju. They appear to be overseeing the smooth operation of the voting process, ensuring its efficient management amidst challenges posed by the damaged building. |
The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea. | [
"\"Severe earthquake hits Gyeongju, South Korea in 2016\"",
"\"Gyeongju earthquake in South Korea in 2016\"",
"\"Gyeongju earthquake in 2016\"",
"\"2016 Gyeongju earthquake\"",
"The Gyeongju 2016 earthquake"
] | [
"Gyeongju earthquake 2016",
"Rescue people affected by the 2016 Gyeongju earthquake",
"\"Gyeongju earthquake rescue efforts in 2016, South Korea\"",
"\"2016 Gyeongju earthquake\" - related to destruction of buildings",
"Rescue efforts in progress Gyeongju, South Korea in 2016"
] | [
"Gyeongju 2016 earthquake",
"South Korea earthquake 2016"
] | {
"article_url": "https://ko.wikipedia.org/wiki/2016%EB%85%84_%EA%B2%BD%EC%A3%BC_%EC%A7%80%EC%A7%84",
"during_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to decompose the ongoing event in multiple simple events. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"during_raw_generation": "EXPLANATION:\nThe given query \"The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea\" refers to a specific event. It does not contain any visual data but includes descriptive language that could be used to visualize the event in a video.\n\nEVENTS:\n1. The 2016 Gyeongju earthquake\n2. South Korea",
"en_article_excerpt": "The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.",
"en_article_url": "https://en.wikipedia.org/wiki/2016_Gyeongju_earthquake",
"event_category": "disasters",
"event_name": "gyeongju_earthquake",
"language": "korean",
"prequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what previous events could lead to this given event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"prequel_raw_generation": "EXPLANATION:\nThis query could be visualized in a video as a news report or a footage of the event. Given the specifics of the event, it's likely that the video will show destruction and damage, which could be reflected in a caption. Therefore, I will focus on extracting events that could be associated with the video.\n\nEVENTS:\n1. Destruction\n2. Damage\n3. Disaster\n4. Earthquake\n5. Rescue",
"sequel_prompt": "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nCutting Knowledge Date: December 2023\nToday Date: 16 Jan 2025\n\n<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nYou are given a video search query. Assume the video search query is an ongoing event. You have to find out what events can be the outcome of this ongoing event. The extracted events should be concrete enough to be visualized in a video.\n\nHere are some guidelines: \n- Extract events that could be visualized in a video.\n- Before extracting an event, consider whether the event text could serve as a reasonable caption for a frame of a video associated with the query.\n- Give specific events related to the query, use your previous WORLD KNOWLEDGE to extract events that are likely to be seen in a video. For example, if the query is \"2019 Christmas Eve Protests in Hong Kong\", you might return \"Protesters marching in the streets\", but not \"Protestors marching in snow\" because it is not likely to have snow in Hong Kong, though its pretty common in other places. \n- First, explain your reasoning in 1-2 sentences about which events it makes sense to extract. Then, provide a numbered list with the events (anywhere from one to five event; use your best judgement to determine how many to extract)\n- Structure your response into two sections, \"\"EXPLANATION:\"\" and then \"\"EVENTS:\"\". The events sections should contain nothing but the events. \n\nHere is the query I'd like you to extract events from: The 2016 Gyeongju earthquake occurred on September 12 near Gyeongju, South Korea.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"sequel_raw_generation": "EXPLANATION:\nGiven the 2016 Gyeongju earthquake, it is a significant historical event with known visual imagery associated with it. Therefore, it is reasonable to extract events that could be seen in a video related to the earthquake, such as scenes of destruction, people fleeing the area, and rescue efforts.\n\nEVENTS:\n1. Destruction\n2. Rescue\n3. Rescue Efforts\n4. Destruction of Buildings\n5. Rescue Efforts in Progress",
"video_URL": "http://youtube.com/watch?v=w6jdwGYcKZU",
"video_description": "안전처 \"경주지진 부상자 23명ㆍ재산피해 1,118건\" 국민안전처는 경주에서 발생한 규모 5.8의 지진으로 부상자 23명이 발생하고, 재산상 피해는 1천118건으로 집계됐다고 밝혔습니다. 안전처와 보건복지부에 따르면 부상자 23명 가운데 7명은 병원에서 치료를 받고 귀가했으며, 16명은 입원 중입니다. 지역별 부상자는 경북이 12명으로 가장 많고, 울산 4명, 대구 2명, 부산·인천·충북·전남·경남 각 1명입니다. 안전처는 균열이 발생한 주택을 대상으로 구조기술사 등과 함께 안전점검을 하고 귀가 여부와 추가 안전조치 등을 검토할 예정입니다. 연합뉴스TV : 02-398-4409(제보) 4441(기사문의), 카톡/라인 jebo23 (끝)",
"video_id": "w6jdwGYcKZU"
} | 16 | [
"A news anchor is delivering a report in front of a cityscape background. The text on the screen indicates a news story about a collision involving a government official, mentioning \"collision\" and \"government official\" along with details about the incident. The temperature and time are also displayed.",
"The... | w6jdwGYcKZU | The video depicts a single news report about an earthquake that occurred in Gyeongju, South Korea. The event began with a collision involving a government official and resulted in 23 injuries and 1,118 property damage cases. The initial scenes show the aftermath of the earthquake, including a snowy and rainy scene, with the government official being treated at a hospital. The footage then shifts to a series of close-ups of the damaged area, including cracked walls, broken furniture, and debris. The earthquake and typhoon ( typhoon "12-1-15") mentioned in the video indicate that the earthquake was one of several devastating natural disasters in South Korea. The damaged areas are shown in the early morning hours, with the focus on the aftermath of the earthquakes and the ongoing efforts to respond to the disaster. |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 6
Size of downloaded dataset files:
8.48 MB
Size of the auto-converted Parquet files:
8.48 MB
Number of rows:
2,394