Dataset Preview
Full Screen Viewer
Full Screen
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 1 new columns ({'caption'}) and 1 missing columns ({'response'}).
This happened while the json dataset builder was generating data using
hf://datasets/lscpku/TimeR-v1/annotations/test/anet_localization.jsonl (at revision ebc1ad52dc59d0d628805de7d415c6d18e183570)
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 2011, in _prepare_split_single
writer.write_table(table)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py", line 585, in write_table
pa_table = table_cast(pa_table, self._schema)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2302, in table_cast
return cast_table_to_schema(table, schema)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2256, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
video_id: string
caption: string
timestamps: list<item: list<item: double>>
child 0, item: list<item: double>
child 0, item: double
instruction: string
question_type: string
to
{'instruction': Value(dtype='string', id=None), 'response': Value(dtype='string', id=None), 'timestamps': Sequence(feature=Sequence(feature=Value(dtype='float64', id=None), length=-1, id=None), length=-1, id=None), 'video_id': Value(dtype='string', id=None), 'question_type': Value(dtype='string', id=None)}
because column names don't match
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1577, in compute_config_parquet_and_info_response
parquet_operations = convert_to_parquet(builder)
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1191, in convert_to_parquet
builder.download_and_prepare(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1027, in download_and_prepare
self._download_and_prepare(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1882, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 2013, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 1 new columns ({'caption'}) and 1 missing columns ({'response'}).
This happened while the json dataset builder was generating data using
hf://datasets/lscpku/TimeR-v1/annotations/test/anet_localization.jsonl (at revision ebc1ad52dc59d0d628805de7d415c6d18e183570)
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
instruction
string | response
string | timestamps
sequence | video_id
string | question_type
string |
|---|---|---|---|---|
Find the segment within the video that addresses the question and formulate a response based on that segment. Format your reply with two components: timestamps and the answer.
Question: Which object did the person interact with after making some food? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Window. | [
[
12.88,
20.29
]
] | 46GP8 | sequence |
Find the segment within the video that addresses the question and formulate a response based on that segment. Give the beginning and ending timestamps, then proceed with your response.
Question: Before watching outside of a window, what was the person holding? | You will find the response to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
Answer: Food. | [
[
3.79,
8.83
]
] | 46GP8 | sequence |
You are required to determine the appropriate segments within the video, indicated by their timestamps, and use the content within those sections to address the given question. Your response should include the starting and ending timestamps, followed by your answer.
Question: In the video, which object were they holding? | The question can be addressed within the timeframe of <start_time_0>s - <end_time_0>s in the video.
The answer is: Food. | [
[
3.79,
8.83
]
] | 46GP8 | description |
Based on the visual content and the following question, first find out which part of the video can be used to answer the question, then answer the question accordingly. Give the beginning and ending timestamps, then proceed with your response.
Question: Before watching outside of a window, which object were they touching? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
Answer: Food. | [
[
3.79,
8.83
]
] | 46GP8 | sequence |
Determine the section of the video relevant to addressing the question and provide an appropriate response. Format your reply with two components: timestamps and the answer.
Question: What did they touch after making some food? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Window. | [
[
12.88,
20.29
]
] | 46GP8 | sequence |
Locate the pertinent segment of the video in relation to the question and provide a relevant answer. Format your reply with two components: timestamps and the answer.
Question: After making some food, what did they go behind? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Window. | [
[
12.88,
18.42
]
] | 46GP8 | sequence |
Find the segment within the video that addresses the question and formulate a response based on that segment. Provide two sets of timestamps - start and end - along with your answer.
Question: Which object were they above before watching outside of a window? | The information needed to answer the question can be found between <start_time_0> and <end_time_0> seconds in the video.
Answer: Food. | [
[
1.29,
6.33
]
] | 46GP8 | sequence |
Determine the section of the video relevant to addressing the question and provide an appropriate response. Your reply should consist of the starting and ending timestamps, followed by your answer.
Question: In the video, which object did the person go above? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
Answer: Food. | [
[
6.33,
8.83
]
] | 46GP8 | description |
Your task involves identifying the relevant timestamps within the video and using the information within those segments to provide an answer to the question. Your reply should include two parts: timestamps and the answer.
Question: After making some food, which object were they on the side of? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Window. | [
[
12.88,
20.29
]
] | 46GP8 | sequence |
Discover the portion of the video that corresponds to the question and deliver a suitable response based on that segment. Your response should be divided into two fields: timestamps and the answer.
Question: Which object were they watching after making some food? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Window. | [
[
12.88,
20.29
]
] | 46GP8 | sequence |
Your goal is to determine the appropriate time points within the video and utilize the content within those segments to address the question provided. Your response should be divided into two fields: timestamps and the answer.
Question: In the video, what did they watch? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Window. | [
[
12.88,
20.29
]
] | 46GP8 | description |
Your goal is to determine the appropriate time points within the video and utilize the content within those segments to address the question provided. Provide two sets of timestamps - start and end - along with your answer.
Question: Before watching outside of a window, what was the person doing? | You will find the response to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
The answer to the question is: Making some food. | [
[
1.29,
11.9
]
] | 46GP8 | sequence |
Determine the section of the video relevant to addressing the question and provide an appropriate response. Your response should be divided into two fields: timestamps and the answer.
Question: After taking a broom from somewhere, what was the person standing on? | Timestamps: (<start_time_0> s, <end_time_0> s)
Answer: Floor. | [
[
9.2,
12.52
]
] | N11GT | sequence |
Find the segment within the video that addresses the question and formulate a response based on that segment. Present the starting and ending timestamps, and then provide your answer.
Question: Which object were they carrying before starting to hold a broom? | Look for the answer to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
The answer to the question is: Blanket. | [
[
1.2,
8.12
]
] | N11GT | sequence |
Your responsibility is to locate the relevant moments within the video, marked by their timestamps, and use the information within those segments to answer the question given. Your response should include the starting and ending timestamps, followed by your answer.
Question: Which object were they holding while taking a broom from somewhere? | The answer to the question is contained within the segment of the video spanning from <start_time_0> seconds to <end_time_0> seconds.
The answer to the question is: Blanket. | [
[
5.88,
8.2
]
] | N11GT | sequence |
You are expected to find the relevant segments within the video, noted by their timestamps, and incorporate the content from those sections to respond to the question. Give the beginning and ending timestamps, then proceed with your response.
Question: What did they hold after tidying something on the floor? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
The answer to the question is: Broom. | [
[
12.52,
13.68
]
] | N11GT | sequence |
Discover the portion of the video that corresponds to the question and deliver a suitable response based on that segment. Structure your response into two fields: timestamps and the answer.
Question: What was the person holding after sneezing somewhere? | Timestamps of relevant clips: <start_time_0> seconds to <end_time_0> seconds
Answer: Broom. | [
[
12.52,
13.68
]
] | N11GT | sequence |
You will need to pinpoint the relevant time frames within the video and draw upon the content within those segments to answer the question presented. Include both the starting and ending timestamps before presenting your answer.
Question: What was the person holding after tidying up a blanket? | The timeframe for answering the question extends from <start_time_0> seconds to <end_time_0> seconds in the video.
The answer is: Broom. | [
[
12.52,
13.68
]
] | N11GT | sequence |
Your responsibility is to locate the relevant moments within the video, marked by their timestamps, and use the information within those segments to answer the question given. Structure your response into two fields: timestamps and the answer.
Question: Before taking a broom from somewhere, what did they hold? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Blanket. | [
[
1.2,
5.88
]
] | N11GT | sequence |
Your task involves identifying the relevant timestamps within the video and using the information within those segments to provide an answer to the question. Your response should be divided into two fields: timestamps and the answer.
Question: Which object did the person hold before starting to sneeze somewhere? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Blanket. | [
[
1.2,
7
]
] | N11GT | sequence |
Your job is to discover the pertinent timestamps within the video and utilize the content within those segments to formulate your response to the question at hand. Format your reply with two components: timestamps and the answer.
Question: Which object were they standing on between sneezing somewhere and putting a broom somewhere? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Floor. | [
[
12.52,
13.64
]
] | N11GT | description |
Locate the specific portion of the video containing the information needed to answer the question, then provide the answer. Your reply should include two parts: timestamps and the answer.
Question: What did they stand on between tidying up a blanket and putting a broom somewhere? | Timestamps of relevant clips: <start_time_0> seconds to <end_time_0> seconds
Answer: Floor. | [
[
12.52,
13.64
]
] | N11GT | description |
Pinpoint the section of the video relevant to the question and supply an answer based on that section. Your response must contain the starting and ending timestamps, followed by your answer.
Question: While taking a broom from somewhere, which object did the person stand on? | Refer to the content between <start_time_0> seconds and <end_time_0> seconds in the video to answer the question.
The answer to the question is: Floor. | [
[
5.88,
8.2
]
] | N11GT | sequence |
Determine the part of the video containing the information needed to answer the question and respond accordingly. Divide your response into two fields: timestamps and the answer.
Question: Which object were they standing on while sneezing somewhere? | Timestamps of relevant clips: <start_time_0> seconds to <end_time_0> seconds
Answer: Floor. | [
[
7,
11.52
]
] | N11GT | sequence |
Your responsibility is to locate the relevant moments within the video, marked by their timestamps, and use the information within those segments to answer the question given. Format your reply with two components: timestamps and the answer.
Question: While tidying up a blanket, what did they stand on? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Floor. | [
[
1.2,
11.52
]
] | N11GT | sequence |
Locate the pertinent segment of the video in relation to the question and provide a relevant answer. Your response should be divided into two fields: timestamps and the answer.
Question: What was the person standing on while holding a broom? | Timestamps of relevant clips: <start_time_0>s - <end_time_0>s
Answer: Floor. | [
[
9.2,
12.52
]
] | N11GT | sequence |
You will need to pinpoint the relevant time frames within the video and draw upon the content within those segments to answer the question presented. Your response should be divided into two fields: timestamps and the answer.
Question: Which object were they standing on? | Timestamps: (<start_time_0> s, <end_time_0> s)
Answer: Floor. | [
[
1.2,
12.52
]
] | N11GT | description |
You will be given a question about the video. Your task is to find out the relevant segments in the video (in second), and respond to the question based on the content of the video. Your response should include the starting and ending timestamps, followed by your answer.
Question: What did they stand on before starting to put a broom somewhere? | The question can be addressed within the timeframe of <start_time_0>s - <end_time_0>s in the video.
The answer is: Floor. | [
[
1.2,
13.68
]
] | N11GT | sequence |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Include both the starting and ending timestamps before presenting your answer.
Question: Which object did the person stand on before putting a broom somewhere but after taking a broom from somewhere? | You will find the response to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
Answer: Floor. | [
[
9.2,
13.64
]
] | N11GT | sequence |
You will be given a question about the video. Your task is to find out the relevant segments in the video (in second), and respond to the question based on the content of the video. Your response should be divided into two fields: timestamps and the answer.
Question: What did they stand on after tidying up a blanket? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Floor. | [
[
12.52,
13.68
]
] | N11GT | sequence |
Based on the visual content and the following question, first find out which part of the video can be used to answer the question, then answer the question accordingly. Provide two sets of timestamps - start and end - along with your answer.
Question: Before taking a broom from somewhere, which object were they standing on? | You will find the response to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
Answer: Floor. | [
[
1.2,
5.88
]
] | N11GT | sequence |
You will be given a question about the video. Your task is to find out the relevant segments in the video (in second), and respond to the question based on the content of the video. Your response must contain the starting and ending timestamps, followed by your answer.
Question: Before starting to sneeze somewhere, what did they stand on? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
Answer: Floor. | [
[
1.2,
7
]
] | N11GT | sequence |
You will need to pinpoint the relevant time frames within the video and draw upon the content within those segments to answer the question presented. Your response should be divided into two fields: timestamps and the answer.
Question: Which object were they standing on before starting to hold a broom? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Floor. | [
[
1.2,
9.2
]
] | N11GT | sequence |
Your responsibility is to locate the relevant moments within the video, marked by their timestamps, and use the information within those segments to answer the question given. Your response should include the starting and ending timestamps, followed by your answer.
Question: Before starting to put a broom somewhere, what was the person carrying? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
The answer is: Blanket. | [
[
8.12,
11.52
]
] | N11GT | sequence |
You will need to pinpoint the relevant time frames within the video and draw upon the content within those segments to answer the question presented. Provide two sets of timestamps - start and end - along with your answer.
Question: In the video, which object were they carrying? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
The answer to the question is: Blanket. | [
[
8.12,
11.52
]
] | N11GT | description |
Your goal is to determine the appropriate time points within the video and utilize the content within those segments to address the question provided. Structure your response into two fields: timestamps and the answer.
Question: After taking a broom from somewhere, what did they carry? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Blanket. | [
[
9.2,
11.52
]
] | N11GT | sequence |
You are expected to find the relevant segments within the video, noted by their timestamps, and incorporate the content from those sections to respond to the question. Your response should be divided into two fields: timestamps and the answer.
Question: What did they put down after sneezing somewhere? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Broom. | [
[
12.52,
13.68
]
] | N11GT | sequence |
Your job is to discover the pertinent timestamps within the video and utilize the content within those segments to formulate your response to the question at hand. Include both the starting and ending timestamps before presenting your answer.
Question: While taking a broom from somewhere, what was the person behind? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
The answer to the question is: Blanket. | [
[
5.88,
8.2
]
] | N11GT | sequence |
Locate the pertinent segment of the video in relation to the question and provide a relevant answer. Format your reply with two components: timestamps and the answer.
Question: After tidying up a blanket, which object did the person go on the side of? | Timestamps: (<start_time_0> s, <end_time_0> s)
Answer: Floor. | [
[
12.52,
13.68
]
] | N11GT | sequence |
You are required to determine the appropriate segments within the video, indicated by their timestamps, and use the content within those sections to address the given question. Present the starting and ending timestamps, and then provide your answer.
Question: What did they carry before putting a broom somewhere but after taking a broom from somewhere? | The information needed to answer the question can be found between <start_time_0> and <end_time_0> seconds in the video.
The answer is: Blanket. | [
[
9.2,
13.64
]
] | N11GT | sequence |
Find the segment within the video that addresses the question and formulate a response based on that segment. Divide your response into two fields: timestamps and the answer.
Question: While tidying something on the floor, what did they carry? | Timestamps: (<start_time_0> s, <end_time_0> s)
Answer: Blanket. | [
[
8.12,
11.52
]
] | N11GT | sequence |
Analyze the video content to identify the segment pertinent to addressing the question, then respond accordingly. Your response should be divided into two fields: timestamps and the answer.
Question: While sneezing somewhere, which object did the person carry? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Blanket. | [
[
8.12,
11.52
]
] | N11GT | sequence |
Pinpoint the section of the video relevant to the question and supply an answer based on that section. Divide your response into two fields: timestamps and the answer.
Question: While holding a broom, which object did the person carry? | Timestamps of relevant clips: <start_time_0>s - <end_time_0>s
Answer: Blanket. | [
[
9.2,
11.52
]
] | N11GT | sequence |
You will be given a question about the video. Your task is to find out the relevant segments in the video (in second), and respond to the question based on the content of the video. Divide your response into two fields: timestamps and the answer.
Question: After tidying something on the floor, what did they put down? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Broom. | [
[
12.52,
13.68
]
] | N11GT | sequence |
Determine the part of the video containing the information needed to answer the question and respond accordingly. Your response must contain the starting and ending timestamps, followed by your answer.
Question: What did they go behind before taking a broom from somewhere? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
The answer is: Blanket. | [
[
1.2,
5.88
]
] | N11GT | sequence |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Format your reply with two components: timestamps and the answer.
Question: Which object were they behind before starting to sneeze somewhere? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Blanket. | [
[
1.2,
7
]
] | N11GT | sequence |
Identify the relevant part of the video in relation to the question and give a corresponding answer. Give the beginning and ending timestamps, then proceed with your response.
Question: Between tidying up a blanket and putting a broom somewhere, which object did the person go on the side of? | Refer to the content between <start_time_0> seconds and <end_time_0> seconds in the video to answer the question.
The answer to the question is: Floor. | [
[
12.52,
13.64
]
] | N11GT | description |
Your goal is to determine the appropriate time points within the video and utilize the content within those segments to address the question provided. Include both the starting and ending timestamps before presenting your answer.
Question: Between sneezing somewhere and putting a broom somewhere, what was the person on the side of? | The question can be answered using content from <start_time_0>s to <end_time_0> s in the video.
The answer is: Floor. | [
[
12.52,
13.64
]
] | N11GT | description |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Your response must contain the starting and ending timestamps, followed by your answer.
Question: Which object did the person take before starting to sneeze somewhere? | The answer to the question is contained within the segment of the video spanning from <start_time_0> seconds to <end_time_0> seconds.
The answer is: Broom. | [
[
5.88,
7
]
] | N11GT | sequence |
Locate the specific portion of the video containing the information needed to answer the question, then provide the answer. Your response must contain the starting and ending timestamps, followed by your answer.
Question: In the video, what did they take? | The question can be addressed within the timeframe of <start_time_0>s - <end_time_0>s in the video.
Answer: Broom. | [
[
5.88,
8.2
]
] | N11GT | description |
Your task involves identifying the relevant timestamps within the video and using the information within those segments to provide an answer to the question. Your reply should consist of the starting and ending timestamps, followed by your answer.
Question: Which object were they taking while tidying up a blanket? | The question can be answered using content from <start_time_0>s to <end_time_0> s in the video.
The answer is: Broom. | [
[
5.88,
8.2
]
] | N11GT | sequence |
Find the segment within the video that addresses the question and formulate a response based on that segment. Structure your response into two fields: timestamps and the answer.
Question: While sneezing somewhere, what was the person taking? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Broom. | [
[
7,
8.2
]
] | N11GT | sequence |
Your goal is to determine the appropriate time points within the video and utilize the content within those segments to address the question provided. Your reply should consist of the starting and ending timestamps, followed by your answer.
Question: What did they take while tidying something on the floor? | The information needed to answer the question can be found between <start_time_0> and <end_time_0> seconds in the video.
Answer: Broom. | [
[
5.88,
8.2
]
] | N11GT | sequence |
Pinpoint the section of the video relevant to the question and supply an answer based on that section. Give the beginning and ending timestamps, then proceed with your response.
Question: Which object did the person put down after tidying up a blanket? | The question can be answered using content from <start_time_0>s to <end_time_0> s in the video.
Answer: Broom. | [
[
12.52,
13.68
]
] | N11GT | sequence |
Based on the visual content and the following question, first find out which part of the video can be used to answer the question, then answer the question accordingly. Present the starting and ending timestamps, and then provide your answer.
Question: After sneezing somewhere, which object were they on the side of? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
The answer is: Floor. | [
[
12.52,
13.68
]
] | N11GT | sequence |
Based on the visual content and the following question, first find out which part of the video can be used to answer the question, then answer the question accordingly. Your response should include the starting and ending timestamps, followed by your answer.
Question: Before taking a phone from somewhere but after closing a door, which object were they putting down? | Refer to the content between <start_time_0> seconds and <end_time_0> seconds in the video to answer the question.
The answer is: Clothes. | [
[
11.81,
14.74
]
] | KRF68 | sequence |
You will be given a question about the video. Your task is to find out the relevant segments in the video (in second), and respond to the question based on the content of the video. Divide your response into two fields: timestamps and the answer.
Question: Before taking a phone from somewhere but after grasping onto a doorknob, which object were they putting down? | Timestamps of relevant clips: <start_time_0> seconds to <end_time_0> seconds
Answer: Clothes. | [
[
13.38,
14.74
]
] | KRF68 | sequence |
Your objective is to identify the relevant timestamps in the video and respond to the question using the information within those segments. Structure your response into two fields: timestamps and the answer.
Question: Which object did the person put down? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Clothes. | [
[
13.38,
14.74
]
] | KRF68 | description |
Your job is to discover the pertinent timestamps within the video and utilize the content within those segments to formulate your response to the question at hand. Your reply should consist of the starting and ending timestamps, followed by your answer.
Question: After running somewhere, what did they put down? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
Answer: Clothes. | [
[
13.38,
14.74
]
] | KRF68 | sequence |
Locate the specific portion of the video containing the information needed to answer the question, then provide the answer. Format your reply with two components: timestamps and the answer.
Question: After grasping onto a doorknob, which object did the person put down? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Clothes. | [
[
13.38,
14.74
]
] | KRF68 | sequence |
You are expected to find the relevant segments within the video, noted by their timestamps, and incorporate the content from those sections to respond to the question. Give the beginning and ending timestamps, then proceed with your response.
Question: Between running somewhere and taking a phone from somewhere, what did they dress? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
The answer is: Clothes. | [
[
13.38,
20.61
]
] | KRF68 | description |
Locate the pertinent segment of the video in relation to the question and provide a relevant answer. Your reply should include two parts: timestamps and the answer.
Question: What did they dress before taking a phone from somewhere but after closing a door? | Timestamps: (<start_time_0> s, <end_time_0> s)
Answer: Clothes. | [
[
11.81,
20.61
]
] | KRF68 | sequence |
You will need to pinpoint the relevant time frames within the video and draw upon the content within those segments to answer the question presented. Divide your response into two fields: timestamps and the answer.
Question: What did they dress before taking a phone from somewhere but after grasping onto a doorknob? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Clothes. | [
[
10.24,
20.61
]
] | KRF68 | sequence |
Your responsibility is to locate the relevant moments within the video, marked by their timestamps, and use the information within those segments to answer the question given. Your reply should consist of the starting and ending timestamps, followed by your answer.
Question: What was the person dressing? | Look for the answer to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
The answer to the question is: Clothes. | [
[
13.38,
24.18
]
] | KRF68 | description |
Analyze the video content to identify the segment pertinent to addressing the question, then respond accordingly. Offer the starting and ending timestamps first, then proceed to answer the question.
Question: What was the person dressing after running somewhere? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
Answer: Clothes. | [
[
10.24,
24.18
]
] | KRF68 | sequence |
Identify the relevant part of the video in relation to the question and give a corresponding answer. Your response should include the starting and ending timestamps, followed by your answer.
Question: Which object did the person dress after closing a door? | The information needed to answer the question can be found between <start_time_0> and <end_time_0> seconds in the video.
Answer: Clothes. | [
[
11.81,
24.18
]
] | KRF68 | sequence |
Pinpoint the section of the video relevant to the question and supply an answer based on that section. Give the beginning and ending timestamps, then proceed with your response.
Question: What was the person interacting with after dressing themselves? | You will find the response to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
The answer to the question is: Phone. | [
[
25.23,
27.29
]
] | KRF68 | sequence |
Based on the visual content and the following question, first find out which part of the video can be used to answer the question, then answer the question accordingly. Structure your response into two fields: timestamps and the answer.
Question: Before taking a phone from somewhere but after closing a door, which object did the person interact with? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Clothes. | [
[
11.81,
20.61
]
] | KRF68 | sequence |
Locate the pertinent segment of the video in relation to the question and provide a relevant answer. Divide your response into two fields: timestamps and the answer.
Question: Which object were they touching after dressing themselves? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Phone. | [
[
25.23,
27.29
]
] | KRF68 | sequence |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Format your reply with two components: timestamps and the answer.
Question: Which object did the person touch before taking a phone from somewhere but after grasping onto a doorknob? | Timestamps of relevant clips: <start_time_0>s - <end_time_0>s
Answer: Clothes. | [
[
10.24,
20.61
]
] | KRF68 | sequence |
Discover the portion of the video that corresponds to the question and deliver a suitable response based on that segment. Offer the starting and ending timestamps first, then proceed to answer the question.
Question: Which object were they touching before taking a phone from somewhere but after closing a door? | You will find the response to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
Answer: Clothes. | [
[
11.81,
20.61
]
] | KRF68 | sequence |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Format your reply with two components: timestamps and the answer.
Question: Which object were they touching between running somewhere and taking a phone from somewhere? | Timestamps: (<start_time_0> s, <end_time_0> s)
Answer: Clothes. | [
[
10.24,
20.61
]
] | KRF68 | description |
You will need to pinpoint the relevant time frames within the video and draw upon the content within those segments to answer the question presented. Present the starting and ending timestamps, and then provide your answer.
Question: What did they hold after dressing themselves? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
Answer: Phone. | [
[
25.23,
27.29
]
] | KRF68 | sequence |
You are expected to find the relevant segments within the video, noted by their timestamps, and incorporate the content from those sections to respond to the question. Your reply should consist of the starting and ending timestamps, followed by your answer.
Question: What did they hold after putting clothes somewhere? | The answer to the question is contained within the segment of the video spanning from <start_time_0> seconds to <end_time_0> seconds.
Answer: Phone. | [
[
17.68,
27.29
]
] | KRF68 | sequence |
Determine the part of the video containing the information needed to answer the question and respond accordingly. Your response should be divided into two fields: timestamps and the answer.
Question: Which object did the person hold before taking a phone from somewhere but after grasping onto a doorknob? | Timestamps of relevant clips: (<start_time_0> s, <end_time_0> s)
Answer: Clothes. | [
[
10.24,
20.61
]
] | KRF68 | sequence |
Your responsibility is to locate the relevant moments within the video, marked by their timestamps, and use the information within those segments to answer the question given. Divide your response into two fields: timestamps and the answer.
Question: Which object were they holding before taking a phone from somewhere but after closing a door? | Timestamps of relevant clips: <start_time_0>s - <end_time_0>s
Answer: Clothes. | [
[
11.81,
20.61
]
] | KRF68 | sequence |
Locate the specific portion of the video containing the information needed to answer the question, then provide the answer. Present the starting and ending timestamps, and then provide your answer.
Question: Between running somewhere and taking a phone from somewhere, which object did the person hold? | The answer to the question is contained within the segment of the video spanning from <start_time_0> seconds to <end_time_0> seconds.
The answer is: Clothes. | [
[
10.24,
12.58
]
] | KRF68 | description |
Your goal is to determine the appropriate time points within the video and utilize the content within those segments to address the question provided. Your response must contain the starting and ending timestamps, followed by your answer.
Question: Which object were they wearing after grasping onto a doorknob? | The question can be answered using content from <start_time_0>s to <end_time_0> s in the video.
The answer to the question is: Clothes. | [
[
10.24,
24.18
]
] | KRF68 | sequence |
Your goal is to determine the appropriate time points within the video and utilize the content within those segments to address the question provided. Format your reply with two components: timestamps and the answer.
Question: After closing a door, which object were they wearing? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Clothes. | [
[
11.81,
24.18
]
] | KRF68 | sequence |
Discover the portion of the video that corresponds to the question and deliver a suitable response based on that segment. Your response must contain the starting and ending timestamps, followed by your answer.
Question: In the video, which object did the person wear? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
The answer is: Clothes. | [
[
12.58,
24.18
]
] | KRF68 | description |
Your objective is to identify the relevant timestamps in the video and respond to the question using the information within those segments. Your response should be divided into two fields: timestamps and the answer.
Question: What was the person taking after grasping onto a doorknob? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Phone. | [
[
10.24,
27.29
]
] | KRF68 | sequence |
You will need to pinpoint the relevant time frames within the video and draw upon the content within those segments to answer the question presented. Structure your response into two fields: timestamps and the answer.
Question: After dressing themselves, what was the person taking? | Timestamps of relevant clips: <start_time_0> seconds to <end_time_0> seconds
Answer: Phone. | [
[
25.23,
27.29
]
] | KRF68 | sequence |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Your reply should include two parts: timestamps and the answer.
Question: Which object did the person take after closing a door? | Timestamps of relevant clips: <start_time_0> seconds to <end_time_0> seconds
Answer: Phone. | [
[
11.81,
27.29
]
] | KRF68 | sequence |
Analyze the video content to identify the segment pertinent to addressing the question, then respond accordingly. Present the starting and ending timestamps, and then provide your answer.
Question: What did they take after running somewhere? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
Answer: Phone. | [
[
10.24,
27.29
]
] | KRF68 | sequence |
Determine the part of the video containing the information needed to answer the question and respond accordingly. Your response must contain the starting and ending timestamps, followed by your answer.
Question: What was the person taking after holding some clothes? | The question can be answered using content from <start_time_0>s to <end_time_0> s in the video.
Answer: Phone. | [
[
13.38,
27.29
]
] | KRF68 | sequence |
Identify the relevant part of the video in relation to the question and give a corresponding answer. Your response must contain the starting and ending timestamps, followed by your answer.
Question: After putting clothes somewhere, what did they take? | The answer to the question is contained within the segment of the video spanning from <start_time_0> seconds to <end_time_0> seconds.
The answer is: Phone. | [
[
23.16,
27.29
]
] | KRF68 | sequence |
Locate the pertinent segment of the video in relation to the question and provide a relevant answer. Provide two sets of timestamps - start and end - along with your answer.
Question: Which object were they taking while grasping onto a doorknob? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
Answer: Clothes. | [
[
6.88,
9.05
]
] | KRF68 | sequence |
Determine the section of the video relevant to addressing the question and provide an appropriate response. Structure your response into two fields: timestamps and the answer.
Question: Which object did the person close before dressing themselves but after taking some clothes from somewhere? | Timestamps: (<start_time_0> s, <end_time_0> s)
Answer: Door. | [
[
10.24,
12.58
]
] | KRF68 | sequence |
Analyze the video content to identify the segment pertinent to addressing the question, then respond accordingly. Divide your response into two fields: timestamps and the answer.
Question: What did they close before starting to dress themselves? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Door. | [
[
4.68,
10.76
]
] | KRF68 | sequence |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Your response must contain the starting and ending timestamps, followed by your answer.
Question: Before taking some clothes from somewhere, what was the person closing? | The relevant information to answer the question lies within the timeframe from <start_time_0> seconds to <end_time_0> seconds in the video.
The answer to the question is: Door. | [
[
4.68,
6.88
]
] | KRF68 | sequence |
Your job is to discover the pertinent timestamps within the video and utilize the content within those segments to formulate your response to the question at hand. Your reply should include two parts: timestamps and the answer.
Question: Before holding some clothes, which object were they closing? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Door. | [
[
4.68,
10.24
]
] | KRF68 | sequence |
Your responsibility is to locate the relevant moments within the video, marked by their timestamps, and use the information within those segments to answer the question given. Format your reply with two components: timestamps and the answer.
Question: Before putting clothes somewhere, which object did the person close? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Door. | [
[
4.68,
10.76
]
] | KRF68 | sequence |
Locate the pertinent segment of the video in relation to the question and provide a relevant answer. Structure your response into two fields: timestamps and the answer.
Question: In the video, what did they close? | Timestamps: <start_time_0>s - <end_time_0>s
Answer: Door. | [
[
4.68,
10.76
]
] | KRF68 | description |
Pinpoint the section of the video relevant to the question and supply an answer based on that section. Include both the starting and ending timestamps before presenting your answer.
Question: While grasping onto a doorknob, which object were they closing? | The question can be addressed within the timeframe of <start_time_0>s - <end_time_0>s in the video.
Answer: Door. | [
[
4.68,
9.05
]
] | KRF68 | sequence |
Locate the specific portion of the video containing the information needed to answer the question, then provide the answer. Give the beginning and ending timestamps, then proceed with your response.
Question: Which object were they closing while taking some clothes from somewhere? | You can find the answer to the question between <start_time_0> seconds and <end_time_0> seconds in the video.
The answer to the question is: Door. | [
[
6.88,
9.05
]
] | KRF68 | sequence |
You are tasked with identifying the relevant time markers in the video and crafting your response to the question based on the content found within those segments. Your reply should consist of the starting and ending timestamps, followed by your answer.
Question: While running somewhere, which object were they closing? | The information needed to answer the question can be found between <start_time_0> and <end_time_0> seconds in the video.
The answer to the question is: Door. | [
[
4.68,
9.05
]
] | KRF68 | sequence |
Determine the part of the video containing the information needed to answer the question and respond accordingly. Your response should include the starting and ending timestamps, followed by your answer.
Question: What was the person closing before taking a phone from somewhere but after grasping onto a doorknob? | The answer to the question is contained within the segment of the video spanning from <start_time_0> seconds to <end_time_0> seconds.
The answer to the question is: Door. | [
[
10.24,
20.61
]
] | KRF68 | sequence |
Discover the portion of the video that corresponds to the question and deliver a suitable response based on that segment. Include both the starting and ending timestamps before presenting your answer.
Question: What was the person closing between grasping onto a doorknob and dressing themselves? | Look for the answer to the question within the timeframe of <start_time_0> to <end_time_0> seconds in the video.
Answer: Door. | [
[
10.24,
12.58
]
] | KRF68 | description |
You will be given a question about the video. Your task is to find out the relevant segments in the video (in second), and respond to the question based on the content of the video. Your response should be divided into two fields: timestamps and the answer.
Question: Which object did the person close between taking some clothes from somewhere and putting clothes somewhere? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Door. | [
[
10.24,
12.58
]
] | KRF68 | description |
You will be given a question about the video. Your task is to find out the relevant segments in the video (in second), and respond to the question based on the content of the video. Divide your response into two fields: timestamps and the answer.
Question: Before starting to take a phone from somewhere, which object were they closing? | Timestamps: <start_time_0> seconds to <end_time_0> seconds
Answer: Door. | [
[
4.68,
10.76
]
] | KRF68 | sequence |
End of preview.
[Work in Progress] We release TimeR-v1, a dataset that aims to advance Video LLMs towards temporally grounded reasoning. We collect existing video-text datasets with timestamp annotations and construct a large-scale high-quality dataset with the help of powerful LLMs. Each instance in our dataset contains a question-answer pair, as well as a list of timestamps denoting the relevant video segments.
- Downloads last month
- 36