Datasets:
question_id string | question_type string | video string | dimension string | question string | answer string | distractor1 string | distractor2 string | distractor3 string | chart_type string | animation_editorial_layer string | chart_reas_type string | alignment_semantic_label string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
001-1 | EM | 1 | Chart Reasoning | What was the global extreme poverty rate in 1800 according to the wooden board? | 85 | scatter | Value | |||||
001-2 | EM | 1 | Chart Perception | Which color is used to represent Africa in the area chart? | blue | area | ||||||
001-3 | MCQ_single | 1 | Chart Reasoning | Is the rate of decline in extreme poverty speeding up or slowing down? | Speeding up | Slowing down | Unchanged | Don't know | line | Trend | ||
001-4 | MCQ_single | 1 | Chart Perception | What is the title of chart on the wooden board structure? | Don't know | 1800 1900 2000 | 0 50 100 | scatter | ||||
001-5 | EM | 1 | Chart Perception | What is the smallest interval between tick marks on the y-axis for the wooden board structure? | 5 | scatter | ||||||
001-6 | MCQ_single | 1 | Chart Perception | What is the y-axis label of the animated income distribution area chart? | Don't know | EXTREME POVERTY | 1$ 10$ 100$ | area | ||||
001-7 | MCQ_single | 1 | Chart Reasoning | When did the peak of area representing Asia cross the extreme poverty line? | 1995 | 1990 | 2000 | 2005 | area | Value | ||
001-8 | MCQ_single | 1 | Animation | When describing the severe economic development gap between the East and the West around the 1950s, which animation was used? | The animation slows down. | Camera slowly zooms in on the red area. | The red area flash on and off in a regular or intermittent way. | The red area is lit brightly. | area | Timeline | ||
001-9 | MCQ_single | 1 | Narrative | What is the most likely purpose of the presenter repeatedly touching and interacting with the board at the beginning of the video? | To introduce the visual encoding of the visualization. | To introduce the concept of the Industrial Revolution about its initial impact on global poverty reduction. | To create a statistic hook, revealing a surprising percentage of 85% extreme poverty in 1800. | To set a high starting point to create suspense for the future trend of quickly dropping after 1990. | ||||
001-10 | MCQ_single | 1 | Narrative | Why does the video introduce the income distribution chart by a world map? | To introduce color encoding. | To establish a high-level perception for the initial situation. | To engage viewers with the geographic diversity of income. | To showcase fantastic tech usage. | ||||
001-11 | MCQ_single | 1 | Narrative | What is the best title of the video? | How To End Poverty in 15 years | The rise of Asia: A 200-Year History | Hans Rosling: The Truth About Population Growth | The End of the Third World: A Data Story | ||||
001-12 | MCQ_single | 1 | Narrative | What is the core idea that the presenter wants to convey through the income distribution area chart? | The gap between East and West is closing. | There are significant economic disparities among different regions within some countries. | Western countries still maintain a significant economic lead. | The war has inflicted heavy damage on the global economy and population. | ||||
001-13 | Open_ended | 1 | Alignment | 1\n00:00:00,560 --> 00:00:09,759\nthis shows the percent in extreme\n\n2\n00:00:03,800 --> 00:00:14,400\npoverty in the world 0% 50% and\n\n3\n00:00:09,759 --> 00:00:18,000\n100% and each uprise here represent 10\n\n4\n00:00:14,400 --> 00:00:21,119\nyears here we are at 1900 and this is\n\n5\n00:00:18,000 --> 00:00:25,... | there changed the whole pattern of the world it became a divided world the world view you grew up with with a rich West here and with poor Africa and Asia here | data insight | ||||||
002-1 | MCQ_single | 2 | Narrative | What is the best title of the video? | The Real Shape of Colorado: Why It Has 697 Sides | Colorado's Geographical Features Explained | The History of U.S. State Boundaries | How Satellites Revolutionized Map-Making | ||||
002-3 | MCQ_multiple | 2 | Animation | When introducing the Colorado at the very start of the video, what animation is used? | camera zooms in; the area is lit up | the area pulses | a magnifier glass appears to magnify the area | map | Camera | Elements of Visualization | |||
003-1 | MCQ_single | 3 | Chart Reasoning | How did the gap between the Europe(yellow bubbles) and other regions change since 1810? | Increased first, then decreased | Decreased first, then increased | Consistently decreased | Consistently increased | bubble | Trend | ||
003-2 | MCQ_single | 3 | Chart Reasoning | Which country was in the front in both lifespan and income in 1948? | USA | UK | Japan | Netherlands | bubble | Rank | ||
003-3 | EM | 3 | Chart Reasoning | Below what age was life expectancy in all countries in 1810? | 40 | bubble | Value | |||||
003-4 | EM | 3 | Chart Perception | What is the label for the horizontal axis of the bubble chart? | income | bubble | ||||||
003-5 | MCQ_single | 3 | Animation | What entrance animation was used for the Shanghai bubble when it was mentioned? | The Shanghai bubble split off from the China bubble | Camera slowly zooms in on Shanghai bubble | Shanghai bubble vibrate | Shanghai bubble expand and contract rhythmically | bubble | Elements of Visualization | ||
003-7 | MCQ_single | 3 | Narrative | Why does the 'rich and healthy' appear in the coordination system? | to introduce the visual encoding | to claim that the west are rich and healthy | to set target for all contries | to show gap between the poor and rich regions | ||||
003-8 | Open_ended | 3 | Alignment | 1\n00:00:04,540 --> 00:00:09,630\nVisualization is right at the heart of my own work - I teach global health\n\n2\n00:00:10,269 --> 00:00:17,429\nAnd I know having the data is not enough. I have to show it in ways people both enjoy and understand\n\n3\n00:00:17,429 --> 00:00:19,429\nNow, I'm going to try something I've... | here come over countries Europe Brown Asia Red Middle East Queens Africa south of sahara blue and the americas yellow | data context | ||||||
003-9 | EM | 3 | Chart Perception | What is the label for the vertical axis of the bubble chart? | lifespan | bubble | ||||||
003-10 | MCQ_single | 3 | Animation | What animation is used when showing the impact of Spanish flu epidemic around 1910s? | animation slows down | camera zooms in | the chart is colored | the bubbles blink | bubble | Timeline | ||
004-1 | MCQ_single | 4 | Animation | What animation is used to highlight China? | camera pivots along vertical plane. | China bar is lit up | camera zooms out | China bar pulses | bar | Camera | ||
004-2 | EM | 4 | Chart Reasoning | What is the difference in total tonnage of newly launched warships between Japan and Russia, in 10,000 tons(rounded to 2 decimal places)? | 1.05 | bar | Difference | |||||
005-2 | EM | 5 | Chart Reasoning | How many China company are displayed in the video? | 2 | Value | ||||||
006-1 | MCQ_single | 6 | Narrative | Why does the video start with a scale model of '1 ton' and then progressively increase to '1000 tons' and '10,000 tons' before showing country data? | To establish a reference that helps viewers intuitively grasp the magnitude of the pork consumption figures for each country. | To introduce the concept of metric conversion in data visualization | To contrast the weight of different types of cargo | To highlight the logistical challenges of transporting pork | ||||
006-2 | EM | 6 | Chart Reasoning | How many tons of pork does one person consume per year? | 1 | infochart | Value | |||||
007-1 | MCQ_single | 7 | Chart Perception | What color represents China in the 'COAL CONSUMPTION IN MIL. TONS' tree map? | red | blue | yellow | green | tree | |||
007-3 | MCQ_single | 7 | Animation | When mentioning that nuclear energy has the lowest ranking in 'DEATH PER ENERGY UNIT GENERATED PER YEAR', what animation was used? | The NUCLEAR cell expanded and contracted once. | The NUCLEAR cell flashed on and off once. | Camera zoomed in on the NUCLEAR cell. | The NUCLEAR cell changed color. | tree | Elements of Visualization | ||
007-4 | EM | 7 | Chart Reasoning | Assuming the scene represents the year 2020, what would be the average coal consumption per plant in million tons, according to the specific figures provided throughout the video? | 2.7 | combined | Value | |||||
007-5 | EM | 7 | Chart Perception | What is the total number of cells in this 'DEATH PER ENERGY UNIT GENERATED PER YEAR' tree map? | 9 | tree | ||||||
007-6 | EM | 7 | Chart Reasoning | What is the projected number of coal plants in China by 2040? | 2000 | area | Value | |||||
007-7 | EM | 7 | Chart Perception | How many ticks are on the x-axis of the COAL PLANTS chart? | 5 | area | ||||||
007-8 | EM | 7 | Chart Reasoning | What is the rank of nuclear energy in terms of death rate per energy unit produced? Answer with a number only. | 9 | tree | Rank | |||||
007-9 | Open_ended | 7 | Alignment | 1\n00:00:00,039 --> 00:00:03,749\nThree reasons why we should continue\nusing nuclear energy.\n\n2\n00:00:04,269 --> 00:00:07,203\nOne: nuclear energy saves lives.\n\n3\n00:00:07,533 --> 00:00:12,862\nIn 2013, a study conducted by NASA found\nthat nuclear energy has prevented\n\n4\n00:00:12,862 --> 00:00:15,302\naround... | nuclear energy ranks last in death per energy unit produced | data insight | ||||||
007-10 | MCQ_single | 7 | Animation | How does the video present the tree map of 'COAL CONSUMPTION IN MIL. TONS' for different countries? | The cells appear one by one from largest to smallest. | The complete tree map fades in. | The cells appear one by one from smallest to largest. | The camera zooms in on the tree map. | tree | Elements added to Visualization | ||
007-11 | MCQ_single | 7 | Narrative | What visualization is presented to illustrate the contrast between public perception and reality? | a juxtaposition of death by car and by plane | a juxtaposition of waste underground and in the sky | a tree map of coal consumption among different countries | comparison of the amounts of thorium, uranium, and coal required to produce the same amount of energy | ||||
007-12 | EM | 7 | Chart Reasoning | What is China's annual coal consumption in billion tons? | 4 | tree | Value | |||||
007-13 | MCQ_single | 7 | Narrative | What is the best title of the video? | 3 Reasons Why Nuclear Energy Is Awesome | 3 Reasons Why Nuclear Energy Is Terrible | Is Nuclear Energy Awesome? A Debate on Safety | A Comparative Analysis of Energy Mortality and CO2 Emissions | ||||
008-2 | EM | 8 | Chart Reasoning | How many sources contributed to the increase of CO2? | 3 | infochart | Aggregation | |||||
009-1 | MCQ_single | 9 | Narrative | What does the metaphor of the submerged globe primarily serve to argue? | It implies that population expansion has gradually exceeded the planet's carrying capacity. | It claims that population growth will inevitably trigger a resource crisis. | It directly links population growth to global sea-level rise, emphasizing the environmental consequences. | It suggests that growth itself is the drowning risk and call for population management. | ||||
009-2 | EM | 9 | Chart Reasoning | From 1804 to 2010, by how many billions did the global population increase? | 6 | map | Difference | |||||
009-4 | MCQ_single | 9 | Narrative | What is the best title of the video? | 7 Billion: How Did We Get So Big So Fast? | The Beaker Visualization: A Cumulative History of Global Population Metrics | Why Populations Explode: Unpacking the Mechanics of the Demographic Transition | From 1 Billion: Tracking the Origin of Human Expansion | ||||
009-5 | MCQ_single | 9 | Chart Perception | Which region is represented by the red liquid in the glasses? | North America | Africa | China | Europe | bar | |||
009-6 | Open_ended | 9 | Alignment | 1\n00:00:08,200 --> 00:00:13,879\n[Insert subtitle here]\n\n2\n00:00:10,080 --> 00:00:17,119\n[Insert subtitle here]\n\n3\n00:00:13,879 --> 00:00:21,199\n[Insert subtitle here]\n\n4\n00:00:17,119 --> 00:00:23,760\nhow did we grow so much so fast say this\n\n5\n00:00:21,199 --> 00:00:26,279\nglass is North America the F... | a thousand years ago there were only a third of a billion people but we were multiplying now There's 7 billion of us | data insight | ||||||
009-7 | MCQ_single | 9 | Animation | In the area chart labeled with 'BILLIONS' and 'YEAR', what animation is used? | The area chart first displays the final climax, then flashes back to replay the entire process. | The area flash on and off in a regular or intermittent way. | Camera slowly zooms in on the area. | The animation slows down. | area | Timeline | ||
010-1 | MCQ_single | 10 | Animation | What animation is used when the line chart reached its peak? | camera switches from a third-person view to a first-person view along the line | camera zooms in | the peak area is lit up | the screen shakes | line | Camera | ||
010-2 | MCQ_single | 10 | Narrative | What is the video's attitude towards the stock market? | neutral | optimistic | pessimistic | cautious | ||||
010-3 | EM | 10 | Chart Perception | What is the latest year labeled on the x-axis of the chart? | 2018 | |||||||
011-1 | MCQ_single | 11 | Chart Perception | What do the light blue dots falling into the map represent? | the poor 99% population | wealth | millionaires | all the population | map | |||
011-2 | MCQ_single | 11 | Narrative | What is the author's attitude to the '99% vs 1%'? | doubt | agree | disagree | objective | ||||
011-3 | Open_ended | 11 | Alignment | 1\n00:00:00,359 --> 00:00:04,759\nwe are the\n\n2\n00:00:01,720 --> 00:00:07,640\n99% it's the rallying Cry of the Occupy\n\n3\n00:00:04,759 --> 00:00:10,320\nWall Street protesters they say that a\n\n4\n00:00:07,640 --> 00:00:13,480\ntiny minority controls America's wealth\n\n5\n00:00:10,320 --> 00:00:16,560\nso how r... | now one in every seven Americans lives below the poverty line that's a record 46.2 million people one in six Americans have no health insurance that's 50 million people of every 17 Americans at least one will be earning below the minimum wage of $7.25 per hour | data insight | ||||||
011-4 | MCQ_single | 11 | Chart Reasoning | How many times larger has the pay for executives at the nation's largest companies become since the 1970s according to the Washinton Post in the line chart? | 6 | 3 | 4 | 5 | line | Difference | ||
011-5 | MCQ_single | 11 | Chart Reasoning | How much of US net worth did the richest 1% of the US population own shown in the map? | a third | two third | a half | 99% | map | Proportion | ||
011-6 | MCQ_single | 11 | Narrative | Why does the 1% vs 99% appear again at the end? | to correct the answer to the question raised at the start | to show that the richest 400 people represent exactly 0.01% of the population | to highlight that the top 1% owns more wealth than the bottom 90% | to indicate that the "99%" movement includes everyone except the ultra-rich | ||||
011-8 | MCQ_single | 11 | Animation | When introducing one in every seven Americans lives below the poverty line, what animation is used? | one of the people falls below a benchmark line | one of the people blinks | one of the people pulses | one of the people is colored a different color | infochart | Elements added to Visualization | ||
011-9 | EM | 11 | Chart Reasoning | What percentage of income share is represented by the top bar for the 'rich' group in the 1993-2000 period? | 10.3 | bar | Value | |||||
012-1 | MCQ_single | 12 | Narrative | What is the best title of the video? | A brief history of America and Cuba | US Foreign Policy in the 20th Century: Lessons from the Caribbean | The Rise and Fall of Fidel Castro: A Cuban Perspective | The Cold War and its Impact on Latin American Democracies | ||||
012-2 | MCQ_single | 12 | Narrative | What does the change in the map's painting style in the 1990s symbolize? | To reflect the time background. | To show the rapid development of computers | To indicate that relations between the US and Cuba have changed because of Microsoft | To introduce more styles to enhance the artistic value of the video | ||||
012-3 | MCQ_single | 12 | Narrative | What is the video's attitude toward the relationship between the United States and Cuba? | hopeful | critical | objective | pessimistic | ||||
013-2 | MCQ_single | 13 | Narrative | What is the best title of the video? | Earth Temperature Timeline | Global Warming Crisis | Climate Change Explained | The History of Earth's Climate | ||||
013-3 | EM | 13 | Chart Reasoning | How many years after the invention of the first reliable thermometer did the U.S. President's Science Advisory Committee issue its major report on carbon dioxide? | 251 | timeline | Difference | |||||
013-4 | EM | 13 | Chart Reasoning | How many times faster is the rate of temperature rise of 1 degree after 2000 than it was in 10000 BC, rounded to the nearest whole number? | 14 | line | Difference | |||||
013-5 | EM | 13 | Chart Perception | How many country legends appear in 16:48? | 24 | line | ||||||
013-6 | Open_ended | 13 | Alignment | 1\n00:00:03,600 --> 00:00:08,800\nwhen experts warn us about something\n\n2\n00:00:06,480 --> 00:00:10,480\nlike they do with climate change they're\n\n3\n00:00:08,800 --> 00:00:14,080\nnever 100\n\n4\n00:00:10,480 --> 00:00:17,920\nsure so how do we know when they're\n\n5\n00:00:14,080 --> 00:00:17,920\nsure enough\n\... | today's rate of warming is of a different magnitude and it started abruptly | data insight | ||||||
013-7 | MCQ_single | 13 | Animation | When comparing the global warming rate of today's and the past, what animation is used? | the compared parts are circled | the camera zooms in | the parts are shaking | one part is moved away for the other parts | line | Elements added to Visualization | ||
014-1 | EM | 14 | Chart Reasoning | How many times over could the $200 billion wealth tax pay for all federal food assistance programs? | 2 | bar | Proportion | |||||
014-2 | MCQ_single | 14 | Chart Perception | What policy plan is associated with the shortest bar in the comparison chart? | Free public college (Sanders plan) | Food assistance (Like food stamps) | Cut child poverty 60% (Bennet-Brown plan) | bar | ||||
014-3 | EM | 14 | Chart Reasoning | What is the difference between the net worth of my doctor and Beyonce in millions? | 349.5 | bar | Difference | |||||
014-5 | EM | 14 | Chart Perception | What is the highest value shown on the y-axis of the initial wealth distribution chart? | 11 | bar | ||||||
014-6 | MCQ_multiple | 14 | Narrative | From what perspectives is the wealth distribution chart explained? | what consist of a wealth bar; where the specific person's net worth is located | what the color of bar means | what the width of bar means | bar | ||||
014-7 | Open_ended | 14 | Alignment | 1\n00:00:08,730 --> 00:00:14,730\nThat includes cars, houses, yachts, jewelry\n— and even cold, hard cash.\n\n2\n00:00:14,730 --> 00:00:18,320\nIt also includes things we owe other people,\nlike student loans.\n\n3\n00:00:20,580 --> 00:00:23,590\nSo let's take a quick tour of American household\nwealth.\n\n4\n00:00:23,... | We can start over here, where we see people who owe way more than they have. As we move further right, we can see | data insight | ||||||
014-8 | EM | 14 | Chart Perception | What is the title of the chart shown at the beginning of the video? | the wealth of every single American household | bar | ||||||
014-10 | MCQ_single | 14 | Animation | When elaborating on net worth of specific people, what animation is used? | annotation is added to the corresponding bar | the corresponding bar changes color | the corresponding bar blinks | camera zooms in on the corresponding bar | bar | Elements added to Visualization | ||
015-2 | EM | 15 | Chart Perception | In what year was this dynamic network initiated? | 1869 | network | ||||||
015-3 | Open_ended | 15 | Alignment | 1\n00:00:01,040 --> 00:00:09,480\n[Music]\n\n2\n00:00:03,410 --> 00:00:11,870\nthis is nature's publication record 150\n\n3\n00:00:09,480 --> 00:00:14,280\nyears of interconnected research\n\n4\n00:00:11,870 --> 00:00:17,910\nrepresented by the majority of nature's\n\n5\n00:00:14,280 --> 00:00:21,539\npapers a snapshot... | each dot is a paper the color represents the field yellow for Earth and space science green for physics and so on | data context | ||||||
015-4 | MCQ_multiple | 15 | Narrative | How does the video help the audience understand the visualization? | It uses a visual hook of significant Nature papers citing network.; It defines the visual elements, stating nodes are papers and colors represent scientific fields like physics and space science. | It introduces the context by showing the 'Nature' logo to establish the source of the data visualization. | It raises a question about the nature of scientific interconnectedness to stimulate the audience's curiosity from the outset. | |||||
016-2 | EM | 16 | Chart Perception | How many countries countries are listed as biggest users of laptop screens? | 3 | bar | ||||||
016-4 | EM | 16 | Chart Reasoning | What percentage of screen usage involves shifting from one screen to another? | 65 | pie | Proportion | |||||
016-5 | EM | 16 | Chart Reasoning | What percentage of people find TV ads most favorable? | 41 | pie | Proportion | |||||
017-1 | MCQ_single | 17 | Narrative | What is the best title of the video? | All student debt in the US | The history of the National Defense Education Act of 1958 | How the space race led to the student loan crisis in America | A comparison of Bernie Sanders' and Elizabeth Warren's debt plans | ||||
017-2 | EM | 17 | Chart Perception | Which age group has the highest concentration of data points in the 'By age' chart? | 30 | bar | ||||||
017-3 | Open_ended | 17 | Alignment | 1\n00:00:10,000 --> 00:00:12,000\nAnd the US feared that they were falling behind.\n\n2\n00:00:12,000 --> 00:00:14,700\nOne of our greatest and most\nglaring deficiencies is the...\n\n3\n00:00:21,800 --> 00:00:27,620\nAmerica newsreels like this one stoked anxieties that the country was wasting its intellectual talents... | And ever since student loans have helped more and more people go to college. But as demand for higher education increased... ...the cost did, too. So students took out bigger loans each year. | data insight | ||||||
017-4 | MCQ_single | 17 | Chart Reasoning | When there were 10 million students enrolled in college, how much is the total cost of college per year? | 8000 | 10000 | 12000 | 6000 | line | Value | ||
017-5 | MCQ_single | 17 | Narrative | Which of the following charts conveys information most directly related to the Soviet rocket launch in the video? | students enrolled in college | total cost of college per year | average loan per year | |||||
017-6 | MCQ_single | 17 | Animation | When mentioning all the student debt Americans currently owe adds up to 1.6 trillion dollars, what animation is used? | The bubble splits into tremendous parts. | Camera zooms in on the number. | The bubble flash on and off in a regular or intermittent way. | The animation suddenly speed up. | bar | Elements of Visualization | ||
017-7 | MCQ_single | 17 | Chart Reasoning | In the 'By education level' chart, what is the the biggest group? | 4-yr college | master | doctorate | less than high school | bar | Extreme | ||
017-8 | EM | 17 | Chart Reasoning | How many students enrolled in college in 2010 in millions? | 18 | line | Value | |||||
017-9 | EM | 17 | Chart Perception | What is the minimum tick value on the x-axis of the student enrollment chart? | 1950 | line | ||||||
018-1 | Open_ended | 18 | Alignment | 1\n00:00:01,120 --> 00:00:07,200\nNathan Adrian's 100 meter freestyle was\n\n2\n00:00:03,640 --> 00:00:10,599\nfast one of the fastest in Olympic\n\n3\n00:00:07,200 --> 00:00:12,559\nhistory but how fast is that what if we\n\n4\n00:00:10,599 --> 00:00:13,839\nwere swimming not just against the other\n\n5\n00:00:12,559 ... | starting in 1956 swimmers set records in eight of the next nine Olympics | data insight | ||||||
018-2 | MCQ_single | 18 | Narrative | What is the best title of the video? | Men's 100-Meter Freestyle - All the Medalists | The Evolution of Olympic Swimming Records | Breaking the Barrier: How Swimmers Got Faster | A History of the 100-Meter Freestyle | ||||
018-3 | MCQ_single | 18 | Narrative | Why does the video use the 2012 London Olympics to introduce its main topic? | Compare Nathan with all the champions in history to introduce the topic. | Highlight the high prestige and value of the London Olympic Games. | Show that the London Olympic champion is highly competitive compared with previous champions. | Arouse the audience's curiosity. | ||||
018-4 | MCQ_single | 18 | Animation | When introducing the whole picture of the swimming lane chart, what animation is used? | move the camera to high-angle like it is installed on a lifting crane | the camera change from a close-up shot to a long shot | decrease the size of a visualization for overview | camera pivots along vertical plane | infochart | Camera | ||
018-5 | EM | 18 | Chart Reasoning | How many meters farther behind a modern winner would the 1896 middle-lane swimmer be compared to the 1928 right-lane swimmer? | 22 | infochart | Difference | |||||
018-6 | EM | 18 | Chart Reasoning | When did the 55-second barrier fall? | 1964 | infochart | Value | |||||
018-7 | EM | 18 | Chart Perception | What is the latest year shown on the horizonton axis? | 2024 | infochart | ||||||
019-1 | EM | 19 | Chart Reasoning | How far back from Bob Beeman's record is the winner in 2008 Beijing? | 2 | infochart | Difference | |||||
019-2 | EM | 19 | Chart Perception | What is the latest year shown on the timeline of Olympic long jump performances? | 2012 | infochart |
DVBench
DVBench is a benchmark for evaluating multimodal large language models on data videos, a storytelling medium that combines dynamic charts with structured narratives.
- 300 data videos
- 1,000 human-verified question-answer pairs
- Five dimensions: Narrative, Animation, Chart Perception, Chart Reasoning, and Alignment
Dataset Structure
Each row contains:
question_id: unique question identifierquestion_type:EM,MCQ_single,MCQ_multiple, orOpen_endedvideo: video identifier matchingvideos/<video>.mp4dimension: evaluation dimensionquestion: question or subtitle-cloze contextanswer: reference answerdistractor1,distractor2,distractor3: distractor options for single-choice and multiple-choice questionschart_type: chart category for visual questions in Animation, Chart Perception, and Chart Reasoninganimation_editorial_layer: editorial-layer category for Animation questionschart_reas_type: data-insight category for Chart Reasoning questionsalignment_semantic_label: semantic category (Data InsightorData Context) for Alignment questions
Loading
from datasets import load_dataset
dataset = load_dataset("BomiaoWang/DVBench", split="test")
print(dataset[0])
Dataset Statistics
(a) Question distribution. Distribution of questions across the five evaluation dimensions. Chart Reasoning receives the largest share because viewers primarily focus on underlying data insights rather than superficial graphical elements, and current models remain more vulnerable in reasoning than in basic perception.
(b) Chart types. Chart-type distribution of visual questions and its dimension-specific breakdown. Visual questions include Animation, Chart Perception, and Chart Reasoning. Bar and line charts are the most prevalent, while the remaining questions cover a diverse long tail of chart types.
(c) Video length and topics. Video lengths range from under 30 seconds to approximately 37 minutes and are grouped as Short (≤3 minutes), Medium (3–6 minutes), and Long (>6 minutes). The videos cover ten real-world topics, with Politics & Society and Economy & Finance as the two largest categories.
Videos and Rights
The videos are third-party works and are not relicensed under the annotation license. Copyright and related rights remain with their creators and rightsholders. Access and use must comply with applicable law, source-platform terms, and any restrictions imposed by the rightsholder.
License
The benchmark annotations and metadata are released under CC BY-NC-SA 4.0. They are intended for non-commercial academic research and evaluation.
Citation
If you find DVBench useful, please cite:
@inproceedings{wang2026dvbench,
title={DVBench: Benchmarking MLLMs for Understanding Dynamic Charts and Narratives in Data Videos},
author={Wang, Bomiao and Shao, Zekai and Lan, Jiexiang and Fu, Xiaoliang and Zeng, Xingchen and Chen, Siming},
booktitle={Findings of the Association for Computational Linguistics: EMNLP 2026},
year={2026}
}
Repository
Code and evaluation utilities: https://github.com/BomiaoWang/DVBench
- Downloads last month
- 41
