Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code:   DatasetGenerationError
Exception:    ArrowCapacityError
Message:      array cannot contain more than 2147483646 bytes, have 2428524125
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1815, in _prepare_split_single
                  for _, table in generator:
                                  ^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 151, in _generate_tables
                  pa_table = paj.read_json(
                             ^^^^^^^^^^^^^^
                File "pyarrow/_json.pyx", line 342, in pyarrow._json.read_json
                File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
                File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
              pyarrow.lib.ArrowCapacityError: array cannot contain more than 2147483646 bytes, have 2428524125
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1339, 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 972, in convert_to_parquet
                  builder.download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 894, in download_and_prepare
                  self._download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 970, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1702, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1858, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

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.

id
string
image
string
conversations
list
is_thai
bool
type
string
svit_detail_distilled_6687
visual_genome/713364.jpg
[ { "from": "human", "value": "<image>\nเห็นอะไรแปลกๆ ในรูปมั้ย" }, { "from": "gpt", "value": "ภาพนี้แสดงให้เห็นกลุ่มคนที่กำลังอยู่ในบรรยากาศของบาร์หรือพื้นที่ชิมไวน์ ภายในห้องที่ตกแต่งอย่างอบอุ่นด้วยแสงสีเหลืองนวลสบายตา ทางด้านซ้ายของภาพมองเห็นหญิงสาวสองคนที่กำลังสนทนากันอยู่ หนึ่งในนั้นสวมเสื้อแ...
true
svit_detail_distilled
svit_complex_81875
visual_genome/285783.jpg
[ { "from": "human", "value": "<image>\nBased on the objects in the scene, what kind of home environment does this child live in?" }, { "from": "gpt", "value": "The presence of a wooden bookcase, a white chair, a wooden floor, and artwork on the shelves suggests a well-kept and organized home envi...
false
svit_complex
llavav1.5_364642
visual_genome/588.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.26, 0.57, 0.45, 0.67]." }, { "from": "gpt", "value": "A dog on the sidewalk." }, { "from": "human", "value": "Please provide a short description for this region: [0.06, 0.41, 0.16, 0.69]." ...
false
llavav1.5
llavav1.5_579917
ocrvqa/9812343148.jpg
[ { "from": "human", "value": "<image>\nWhat is the title of this book?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Seychelles and Mahe Insight Pocket Guide" }, { "from": "human", "value": "What type of book is this?" }, { "from": "gpt", ...
false
llavav1.5
llavav1.5_242940
coco_2017/000000183982.jpg
[ { "from": "human", "value": "<image>\nWhere is this located?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Slope" } ]
false
llavav1.5
svit_detail_distilled_18248
visual_genome/2324482.jpg
[ { "from": "human", "value": "<image>\nช่วยเล่ารายละเอียดทุกอย่างในรูปนี้ให้ฟังหน่อย" }, { "from": "gpt", "value": "ภาพนี้เป็นภาพของห้องตรวจในคลินิกหรือโรงพยาบาลที่มีความสะอาดและเป็นระเบียบเรียบร้อย ด้านขวามือของภาพมีเตียงตรวจที่ปูด้วยผ้าขาวสะอาดและมีการปรับระดับได้ พื้นที่โดยรอบเตียงนั้นมีลักษณะ...
true
svit_detail_distilled
svit_referring_157636
visual_genome/2322725.jpg
[ { "from": "human", "value": "<image>\nWhat is the location of the wine glass, and what does it contain? Provide the bounding boxes of the mentioned objects." }, { "from": "gpt", "value": "The wine glass [0.722, 0.588, 0.912, 0.858] is located on the table, and it contains red wine [0.766, 0.698,...
false
svit_referring
llavav1.5_413905
visual_genome/2356338.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: Blue bikini bottom." }, { "from": "gpt", "value": "[0.14, 0.44, 0.32, 0.53]" }, { "from": "human", "value": "Please provide a short description for this region: [0....
false
llavav1.5
llavav1.5_219095
coco_2017/000000095711.jpg
[ { "from": "human", "value": "<image>\nDid the player touch the home plate?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "No" }, { "from": "human", "value": "Who is touching home?" }, { "from": "gpt", "value": "Francoeur" }, { "...
false
llavav1.5
llavav1.5_379358
visual_genome/2402820.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: Slices of tomatoes on the pizza." }, { "from": "gpt", "value": "[0.5, 0.57, 0.83, 0.79]" }, { "from": "human", "value": "Please provide a short description for this...
false
llavav1.5
llavav1.5_279450
coco_2017/000000174721.jpg
[ { "from": "human", "value": "<image>\nWhat word is missing from the phrase that ends in America?\nA. wonderful\nB. productive\nC. hello\nD. good\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "D" } ]
false
llavav1.5
llavav1.5_220101
coco_2017/000000228407.jpg
[ { "from": "human", "value": "<image>\nWhat is in the horizon?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Mountains" }, { "from": "human", "value": "What route is this?" }, { "from": "gpt", "value": "5" }, { "from": "human", ...
false
llavav1.5
llavav1.5_620405
open_images/255f659666e2c65b.jpg
[ { "from": "human", "value": "<image>\nProvide a one-sentence caption for the provided image.\nReference OCR token: LIVE, 4:20, PM, BALTIMORE, RENEWING, AMERICA'S, CN, TAKE, TRAIN, RALLY, PHOTOS,, VIDEO, THE, INAUGURATION, EMAIL, THEM, TO, IREPORT®, CNN, .com, BARACK" }, { "from": "gpt", "value":...
false
llavav1.5
llavav1.5_613087
open_images/21d8e18e1fbb06cd.jpg
[ { "from": "human", "value": "<image>\nProvide a one-sentence caption for the provided image.\nReference OCR token: Nikon, TOOD, Nikon, D7000" }, { "from": "gpt", "value": "Nikon D7000 is labeled on this sealed, cardboard box." } ]
false
llavav1.5
llavav1.5_282421
coco_2017/000000372554.jpg
[ { "from": "human", "value": "<image>\nWhat does the green stuff here need?\nA. water\nB. salt\nC. gas\nD. electricity\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "A" } ]
false
llavav1.5
llavav1.5_282094
coco_2017/000000555639.jpg
[ { "from": "human", "value": "<image>\nWhat game is being played here?\nA. golf\nB. frisbee golf\nC. ultimate frisbee\nD. chess\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "C" } ]
false
llavav1.5
llavav1.5_540182
ocrvqa/1574216236.jpg
[ { "from": "human", "value": "<image>\nWhat is the title of this book?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Sewing Leather Accessories: How to Make Custom Belts, Gloves, and Clutches" }, { "from": "human", "value": "What type of book is th...
false
llavav1.5
llavav1.5_257029
coco_2017/000000483475.jpg
[ { "from": "human", "value": "<image>\nWhat is the man to the right directing?\nA. sheep\nB. planes\nC. dogs\nD. cars\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "B" } ]
false
llavav1.5
svit_detail_71996
visual_genome/2381725.jpg
[ { "from": "human", "value": "<image>\nDelve into the particulars of the image." }, { "from": "gpt", "value": "The image showcases a person's open, outstretched white hand, visibly displaying a sugary, glazed doughnut. The fingers are widely spread, and the irregularly shaped doughnut, not entire...
false
svit_detail
wizardlm_0
null
[ { "from": "human", "value": "Generate a grammatically correct sentence using homophones for the following words: 'sight', 'cite', and 'site'." }, { "from": "gpt", "value": "I need to cite the source of the information I found on the site, as it is important to have a reliable source in sight." ...
false
wizardlm
llavav1.5_438314
visual_genome/2325719.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: a small green candle." }, { "from": "gpt", "value": "[0.0, 0.37, 0.06, 0.45]" }, { "from": "human", "value": "Please provide a short description for this region: [0...
false
llavav1.5
svit_conversation_315792
visual_genome/2417101.jpg
[ { "from": "human", "value": "<image>\nWhat's the main activity happening in the image?" }, { "from": "gpt", "value": "The main activity happening in the image is a man catching a frisbee in a field, possibly during a game of Frisbee Golf." }, { "from": "human", "value": "Is the man a...
false
svit_conversation
llavav1.5_277101
coco_2017/000000004984.jpg
[ { "from": "human", "value": "<image>\nWhat force is causing the boards to accelerate forward?\nA. contact force\nB. inertia\nC. kinetic force\nD. friction\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "A" }, { "from": "human", "value"...
false
llavav1.5
llavav1.5_439131
visual_genome/2324711.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.74, 0.33, 0.82, 0.44]." }, { "from": "gpt", "value": "The tower has a weather vane on top." }, { "from": "human", "value": "Please provide the bounding box coordinate of the region this sente...
false
llavav1.5
wizardlm_0
null
[ { "from": "human", "value": "Assign each of the following items to one of three categories: food, inorganic, or living.\nRock" }, { "from": "gpt", "value": "Inorganic." } ]
false
wizardlm
llavav1.5_471654
visual_genome/2369181.jpg
[ { "from": "human", "value": "<image>\nOn which side of the picture is the small food?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Right" }, { "from": "human", "value": "What is the name of the white vegetable?" }, { "from": "gpt", "v...
false
llavav1.5
llavav1.5_488449
visual_genome/2395255.jpg
[ { "from": "human", "value": "<image>\nDoes the boat that is to the left of the man look little and black?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "No" }, { "from": "human", "value": "Are there any boats in the photo that are not white?" }, ...
false
llavav1.5
llavav1.5_359027
coco_2017/000000202146.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: a woman wearing a black shirt." }, { "from": "gpt", "value": "[0.0, 0.16, 0.56, 0.42]" }, { "from": "human", "value": "Please provide the bounding box coordinate of...
false
llavav1.5
llavav1.5_348559
coco_2017/000000025535.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: burnt piece of broccolli." }, { "from": "gpt", "value": "[0.03, 0.12, 0.4, 0.32]" }, { "from": "human", "value": "Please provide a short description for this region...
false
llavav1.5
svit_conversation_3308
visual_genome/1121.jpg
[ { "from": "human", "value": "<image>\nCan you describe the surroundings in the image?" }, { "from": "gpt", "value": "The surroundings suggest that the image is set in a dental clinic. There's a wall with gray tiles and a closed green door. There's also a round metal spit sink for rinsing and spi...
false
svit_conversation
llavav1.5_206331
coco_2017/000000336966.jpg
[ { "from": "human", "value": "<image>\nAre there people in the plane?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "No" }, { "from": "human", "value": "Is the plane old?" }, { "from": "gpt", "value": "No" }, { "from": "human", ...
false
llavav1.5
wizardlm_0
null
[ { "from": "human", "value": "Suppose you have a set of positive words {love, happy, good} and a set of negative words {hate, sad, bad}. For a given sentence, you want to calculate the sentiment score based on the frequency of positive and negative words used. Here's the formula to calculate the score:\nSent...
false
wizardlm
llavav1.5_391759
visual_genome/2386106.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: a white frisbee being caught." }, { "from": "gpt", "value": "[0.34, 0.02, 0.63, 0.12]" }, { "from": "human", "value": "Please provide the bounding box coordinate of...
false
llavav1.5
llavav1.5_583335
ocrvqa/226391183.jpg
[ { "from": "human", "value": "<image>\nWho is the author of this book?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Harry V. Jaffa" }, { "from": "human", "value": "What is the title of this book?" }, { "from": "gpt", "value": "Crisis o...
false
llavav1.5
llavav1.5_467411
visual_genome/2396191.jpg
[ { "from": "human", "value": "<image>\nWhat food is cooked?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Donuts" }, { "from": "human", "value": "Are the doughnuts on the plate white and cooked?" }, { "from": "gpt", "value": "No" }, ...
false
llavav1.5
llavav1.5_173776
coco_2017/000000155450.jpg
[ { "from": "human", "value": "<image>\nWhat color is the person hat?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "White" }, { "from": "human", "value": "Is this a tennis court?" }, { "from": "gpt", "value": "Yes" }, { "from": "...
false
llavav1.5
wizardlm_0
null
[ { "from": "human", "value": "How can I use C# to provide a customizable and inclusive refreshment system for a party that can accommodate various dietary needs and preferences?\nHere is a sample code snippet that can help achieve this goal:\n// Define a dictionary to store dietary restrictions and their cor...
false
wizardlm
llavav1.5_239491
coco_2017/000000129407.jpg
[ { "from": "human", "value": "<image>\nAre the flowers growing?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Yes" }, { "from": "human", "value": "What color is the plant?" }, { "from": "gpt", "value": "Green" }, { "from": "huma...
false
llavav1.5
llavav1.5_164198
coco_2017/000000009866.jpg
[ { "from": "human", "value": "<image>\nAre these fruits tasty?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Yes" }, { "from": "human", "value": "Is there a tablecloth under the bowl of oranges?" }, { "from": "gpt", "value": "Yes" }, ...
false
llavav1.5
svit_detail_35634
visual_genome/2343178.jpg
[ { "from": "human", "value": "<image>\nCan you provide an intricate depiction of the image?" }, { "from": "gpt", "value": "The image showcases a serene beach setting with the clear blue waters of the ocean as the backdrop. The sky is clear, with just a few partial clouds. The beach has white sand...
false
svit_detail
svit_detail_49005
visual_genome/2357376.jpg
[ { "from": "human", "value": "<image>\nCan you dissect the image and describe each element in detail?" }, { "from": "gpt", "value": "This is a dynamic scene at an outdoor train station. A classic steam train, notable for its black and red colors, is pulling into the station. The train is painted ...
false
svit_detail
svit_detail_27661
visual_genome/2334622.jpg
[ { "from": "human", "value": "<image>\nProvide a comprehensive overview of the image." }, { "from": "gpt", "value": "This image portrays a joyful scene set in a public park. The picture is dominated by three children, each engaging with their toy cell phones, sitting in an area rich with lush gre...
false
svit_detail
llavav1.5_475368
visual_genome/2329159.jpg
[ { "from": "human", "value": "<image>\nWhat is pushed by the girl?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Stroller" }, { "from": "human", "value": "What color is the shirt the lady is wearing?" }, { "from": "gpt", "value": "Green...
false
llavav1.5
svit_referring_181323
visual_genome/2325364.jpg
[ { "from": "human", "value": "<image>\nCan you describe the appearance of the windows in the image and their location? Include the coordinates for each mentioned object." }, { "from": "gpt", "value": "There are large glass windows [0.674, 0.224, 0.996, 0.774] in the image. These are on the buildi...
false
svit_referring
llavav1.5_360845
coco_2017/000000319155.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.55, 0.6, 0.78, 0.74]." }, { "from": "gpt", "value": "Smaller oranges to the right of the black thing in middle." }, { "from": "human", "value": "Please provide a short description for this re...
false
llavav1.5
llavav1.5_554683
ocrvqa/415167388.jpg
[ { "from": "human", "value": "<image>\nWhat is the title of this book?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Adaptations: From Text to Screen, Screen to Text" }, { "from": "human", "value": "What is the genre of this book?" }, { "fr...
false
llavav1.5
llavav1.5_211696
coco_2017/000000345786.jpg
[ { "from": "human", "value": "<image>\nAre the lights working?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Yes" }, { "from": "human", "value": "Is the pic very visible?" }, { "from": "gpt", "value": "Yes" }, { "from": "human",...
false
llavav1.5
svit_detail_52463
visual_genome/2361033.jpg
[ { "from": "human", "value": "<image>\nCan you detail the contents of the image?" }, { "from": "gpt", "value": "This is a daytime scene set in a spacious parking lot. The sky overhead is a delightful blue, populated with fluffy white clouds scattered about. Some of the clouds are larger and darke...
false
svit_detail
wizardlm_0
null
[ { "from": "human", "value": "Can you solve the workplace diversity puzzle? Using the data provided in the [Excel table], analyze the statistics to explain the correlation between race, gender, and age with income and job opportunities. It's no secret that discrimination exists in the workplace, but can you ...
false
wizardlm
llavav1.5_441678
visual_genome/2321642.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: The basket is white in color." }, { "from": "gpt", "value": "[0.64, 0.58, 1.0, 0.73]" }, { "from": "human", "value": "Please provide a short description for this re...
false
llavav1.5
svit_detail_4663
visual_genome/4829.jpg
[ { "from": "human", "value": "<image>\nCould you offer an in-depth analysis of the image?" }, { "from": "gpt", "value": "The image is of a well-organized counter, teeming with various dishes, cups, and glasses. The counter, running from top to bottom and nearly filling the frame, is predominantly...
false
svit_detail
llavav1.5_216398
coco_2017/000000058911.jpg
[ { "from": "human", "value": "<image>\nเห็นได้กี่คน?\nตอบคำถามโดยใช้คำหรือวลีเดียว" }, { "from": "gpt", "value": "1" }, { "from": "human", "value": "ประตูเปิดอยู่หรือไม่?" }, { "from": "gpt", "value": "ใช่" }, { "from": "human", "value": "ผู้ชายกำลังทำอะไรอยู่?...
true
llavav1.5
llavav1.5_314637
coco_2017/000000021036.jpg
[ { "from": "human", "value": "<image>\nWhat does the blue float look like?\nA. lobster\nB. horse\nC. cow\nD. egg\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "A" } ]
false
llavav1.5
svit_conversation_292768
visual_genome/2408957.jpg
[ { "from": "human", "value": "<image>\nCan you describe the laptop the woman is using?" }, { "from": "gpt", "value": "The laptop is a mix of silver and black. It seems to be of a larger size and the woman is actively using it, possibly working or communicating." }, { "from": "human", ...
false
svit_conversation
svit_referring_767892
visual_genome/2390939.jpg
[ { "from": "human", "value": "<image>\nWhat can be inferred about the setting given the glass base [0.33, 0.126, 0.516, 0.186] and the fancy silver fork [0.228, 0.188, 0.858, 0.338]? Provide the bounding boxes of the mentioned objects." }, { "from": "gpt", "value": "The presence of a glass base [...
false
svit_referring
llavav1.5_256082
coco_2017/000000447147.jpg
[ { "from": "human", "value": "<image>\nWhat is safest to touch without being burned?\nA. inside stove\nB. food\nC. pan\nD. foil\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "D" } ]
false
llavav1.5
svit_referring_122249
visual_genome/2318768.jpg
[ { "from": "human", "value": "<image>\nCan you describe the lady [0.046, 0.286, 0.23, 0.646] and her relationship to the man [0.336, 0.09, 0.604, 0.934]? Include the coordinates for each mentioned object." }, { "from": "gpt", "value": "The lady [0.046, 0.286, 0.23, 0.646] is walking next to the m...
false
svit_referring
llavav1.5_353460
coco_2017/000000220284.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.22, 0.25, 0.89, 0.75]." }, { "from": "gpt", "value": "White shirt." }, { "from": "human", "value": "Please provide the bounding box coordinate of the region this sentence describes: green shi...
false
llavav1.5
llavav1.5_403930
visual_genome/2369662.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: a black and red umbrella." }, { "from": "gpt", "value": "[0.09, 0.3, 0.23, 0.36]" }, { "from": "human", "value": "Please provide a short description for this region...
false
llavav1.5
wizardlm_0
null
[ { "from": "human", "value": "Can you provide a JSON data of the recipe for making Fajitas with an additional constraint that the dish must be suitable for vegans? Include all the necessary ingredients and their quantities needed to make this dish." }, { "from": "gpt", "value": "Sure, here's a JS...
false
wizardlm
llavav1.5_274563
coco_2017/000000039078.jpg
[ { "from": "human", "value": "<image>\nThis will take you to what metropolis?\nA. belfast\nB. cork\nC. dublin\nD. limerick\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "C" } ]
false
llavav1.5
wizardlm_0
null
[ { "from": "human", "value": "Construct a system of equations with 3 variables that has a unique solution where the sum of the variables equals 8. Provide the solution set for the system." }, { "from": "gpt", "value": "One possible system of equations with 3 variables that has a unique solution w...
false
wizardlm
llavav1.5_227714
coco_2017/000000240967.jpg
[ { "from": "human", "value": "<image>\nWhat is the make and model of the truck shown in the picture?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Dodge ram" }, { "from": "human", "value": "Are the mountains snowy?" }, { "from": "gpt", ...
false
llavav1.5
llavav1.5_284331
coco_2017/000000157933.jpg
[ { "from": "human", "value": "<image>\nWhat products are being displayed?\nA. landline phones\nB. mobile phones\nC. remote controls\nD. calculators\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "B" } ]
false
llavav1.5
llavav1.5_529603
ocrvqa/1476711429.jpg
[ { "from": "human", "value": "<image>\nผู้แต่งหนังสือเล่มนี้คือใคร?\nตอบคำถามโดยใช้คำหรือวลีเดียว" }, { "from": "gpt", "value": "คาร์เตอร์ เพย์ซิงเกอร์" }, { "from": "human", "value": "หนังสือเล่มนี้มีชื่อว่าอะไร?" }, { "from": "gpt", "value": "ที่ซึ่งชายคนหนึ่งยืนอยู่: สอ...
true
llavav1.5
llavav1.5_493526
visual_genome/2331297.jpg
[ { "from": "human", "value": "<image>\nHow is the weather?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Cloudless" }, { "from": "human", "value": "Is the weather cloudy?" }, { "from": "gpt", "value": "No" }, { "from": "human", ...
false
llavav1.5
llavav1.5_226440
coco_2017/000000369936.jpg
[ { "from": "human", "value": "<image>\nWhere is the boy?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "On bed" }, { "from": "human", "value": "Where is the boy sitting?" }, { "from": "gpt", "value": "Bed" }, { "from": "human", ...
false
llavav1.5
llavav1.5_460329
visual_genome/2355950.jpg
[ { "from": "human", "value": "<image>\nWho is wearing the hat?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Man" }, { "from": "human", "value": "Are there any girls to the right of the man that is wearing pants?" }, { "from": "gpt", "v...
false
llavav1.5
llavav1.5_547349
ocrvqa/161145042X.jpg
[ { "from": "human", "value": "<image>\nWho is the author of this book?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "John Lawrence Reynolds" }, { "from": "human", "value": "What is the title of this book?" }, { "from": "gpt", "value": "...
false
llavav1.5
svit_detail_100196
visual_genome/2411516.jpg
[ { "from": "human", "value": "<image>\nCan you portray the image in words?" }, { "from": "gpt", "value": "The image depicts an intriguing scene on what appears to be a sandy path. In the foreground, a woman is riding a bike, taking center stage in the image. She is stylishly attired in brown cord...
false
svit_detail
svit_detail_14731
visual_genome/2320733.jpg
[ { "from": "human", "value": "<image>\nProvide a comprehensive overview of the image." }, { "from": "gpt", "value": "In this vivid image, a young woman is at the center of attention as she prepares to play tennis. She is intensely focused, her gaze determinedly crossing the court. Dressed in a wh...
false
svit_detail
llavav1.5_354188
coco_2017/000000518117.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.16, 0.03, 0.49, 1.0]." }, { "from": "gpt", "value": "Chef with bowl." }, { "from": "human", "value": "Please provide the bounding box coordinate of the region this sentence describes: chef on...
false
llavav1.5
svit_detail_distilled_57295
visual_genome/2366152.jpg
[ { "from": "human", "value": "<image>\nมีอะไรน่าสนใจในรูปบ้าง" }, { "from": "gpt", "value": "ภาพนี้เป็นภาพนกนางนวลที่ยืนอยู่บนชายหาด โดยมีฉากหลังเป็นท้องทะเลที่ส่องประกายแสงระยิบระยับจากแสงแดดที่ตกกระทบผิวน้ำ นกนางนวลตัวนี้มีสีขาวสะอาดตัดกับสีเทาอ่อนที่ปีกและหาง ดูแล้วให้ความรู้สึกสงบและงดงาม\n\n...
true
svit_detail_distilled
llavav1.5_622193
open_images/337e5a8bea655894.jpg
[ { "from": "human", "value": "<image>\nProvide a one-sentence caption for the provided image.\nReference OCR token: 58, SAMSUNG" }, { "from": "gpt", "value": "a television with the word samsung on the television." } ]
false
llavav1.5
llavav1.5_474791
visual_genome/2345025.jpg
[ { "from": "human", "value": "<image>\nWhat is the cellphone on?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Blanket" }, { "from": "human", "value": "What device is on the blanket?" }, { "from": "gpt", "value": "Cell phone" }, { ...
false
llavav1.5
svit_detail_34965
visual_genome/2342467.jpg
[ { "from": "human", "value": "<image>\nGo into detail about the different elements of the image." }, { "from": "gpt", "value": "The scene before you features a freight train, particularly a red boxcar that's in the foreground. A single car of the train stands out in a striking shade of red, almos...
false
svit_detail
svit_detail_33546
visual_genome/2340952.jpg
[ { "from": "human", "value": "<image>\nNarrate what you see in the image in depth." }, { "from": "gpt", "value": "The image presents a close-up view of a mouth-watering, fully-loaded hamburger. The burger features a number of toppings, the most eye-catching of which is a sunny-side-up egg, nestle...
false
svit_detail
llavav1.5_474556
visual_genome/2390630.jpg
[ { "from": "human", "value": "<image>\nWhat color are the birds that are perched on the pole?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Black" }, { "from": "human", "value": "What animal is perched on the pole?" }, { "from": "gpt", ...
false
llavav1.5
llavav1.5_440630
visual_genome/2322930.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.73, 0.34, 0.89, 0.43]." }, { "from": "gpt", "value": "A curve on the side of the plane." }, { "from": "human", "value": "Please provide the bounding box coordinate of the region this sentence...
false
llavav1.5
svit_referring_504300
visual_genome/2361639.jpg
[ { "from": "human", "value": "<image>\nWhat can be observed about the material used for the headboard [0.146, 0.376, 0.366, 0.46] and the bedspread [0.15, 0.392, 0.794, 0.684]? Provide the bounding boxes of the mentioned objects." }, { "from": "gpt", "value": "The headboard [0.146, 0.376, 0.366, ...
false
svit_referring
llavav1.5_335837
coco_2017/000000061184.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: the couch in the bottom left." }, { "from": "gpt", "value": "[0.04, 0.61, 0.26, 0.8]" } ]
false
llavav1.5
llavav1.5_466746
visual_genome/2357293.jpg
[ { "from": "human", "value": "<image>\nOn which side of the picture is the catcher?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Right" }, { "from": "human", "value": "Is it an indoors scene?" }, { "from": "gpt", "value": "No" }, {...
false
llavav1.5
svit_conversation_297094
visual_genome/2410458.jpg
[ { "from": "human", "value": "<image>\nWhat is the environment like where the surfer is?" }, { "from": "gpt", "value": "The surfer is in the ocean. The water appears to be stirred up and foamy, suggesting that the sea conditions are suitable for surfing." }, { "from": "human", "value"...
false
svit_conversation
llavav1.5_407628
visual_genome/2364694.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.5, 0.4, 0.53, 0.43]." }, { "from": "gpt", "value": "The giraffe's eye is black." }, { "from": "human", "value": "Please provide the bounding box coordinate of the region this sentence describ...
false
llavav1.5
llavav1.5_312623
coco_2017/000000087862.jpg
[ { "from": "human", "value": "<image>\nWhy are the people in the stands?\nA. working\nB. watching\nC. sleeping\nD. playing\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "B" } ]
false
llavav1.5
llavav1.5_243236
coco_2017/000000415728.jpg
[ { "from": "human", "value": "<image>\nนักเขียนแฟนตาซีชื่อดังคนไหนที่มีชื่อเกือบจะเหมือนกันกับนักเบสบอลคนนี้?\nตอบคำถามโดยใช้คำหรือวลีเดียว" }, { "from": "gpt", "value": "ทาด วิลเลียมส์" } ]
true
llavav1.5
llavav1.5_286774
coco_2017/000000312050.jpg
[ { "from": "human", "value": "<image>\nWhy are the people standing behind the black netting?\nA. to wrestle\nB. to socialize\nC. to compete\nD. watching game\nAnswer with the option's letter from the given choices directly." }, { "from": "gpt", "value": "D" } ]
false
llavav1.5
llavav1.5_488505
visual_genome/2390028.jpg
[ { "from": "human", "value": "<image>\nIs the dog that looks brown carrying a frisbee?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Yes" }, { "from": "human", "value": "What animal is playing in the river?" }, { "from": "gpt", "value":...
false
llavav1.5
svit_detail_distilled_84161
visual_genome/2394572.jpg
[ { "from": "human", "value": "<image>\nกรุณาให้คำอธิบายที่ละเอียดถี่ถ้วนเกี่ยวกับภาพนี้" }, { "from": "gpt", "value": "ในภาพนี้มีฉากที่น่าสนใจและเต็มไปด้วยรายละเอียดมากมาย ในภาพนี้เราเห็นคนสองคนเป็นตัวละครหลัก คนหนึ่งเป็นตัวตลกที่นั่งอยู่บนจักรยานยนต์ขนาดเล็กโดยสวมเสื้อผ้าลายทางและสวมหมวกขนาดใหญ่...
true
svit_detail_distilled
svit_detail_distilled_94546
visual_genome/2405555.jpg
[ { "from": "human", "value": "<image>\nกรุณาถ่ายทอดรายละเอียดของภาพนี้เป็นคำพูด" }, { "from": "gpt", "value": "ในภาพนี้แสดงให้เห็นถึงฉากของชายหาดที่สงบและกว้างขวาง มีชายขี่ม้าสีดำอยู่บริเวณกลางภาพ ม้าตัวนี้มีขนสีดำสนิทและดูแข็งแรง มันกำลังเดินไปตามชายหาด ซึ่งมีคลื่นทะเลเบาๆ ซัดเข้ามาบริเวณฝั่ง ชา...
true
svit_detail_distilled
llavav1.5_336226
coco_2017/000000037059.jpg
[ { "from": "human", "value": "<image>\nPlease provide the bounding box coordinate of the region this sentence describes: a man siting on a chair near the bed where a black dog with a cap." }, { "from": "gpt", "value": "[0.64, 0.13, 0.91, 0.47]" }, { "from": "human", "value": "Please p...
false
llavav1.5
llavav1.5_423730
visual_genome/2343407.jpg
[ { "from": "human", "value": "<image>\nPlease provide a short description for this region: [0.4, 0.29, 0.59, 0.46]." }, { "from": "gpt", "value": "Man is wearing all black." }, { "from": "human", "value": "Please provide a short description for this region: [0.46, 0.29, 0.49, 0.33]." ...
false
llavav1.5
wizardlm_0
null
[ { "from": "human", "value": "How can I create a Java program to simulate a game of rock-paper-scissors? Here's the basic outline of what I want to accomplish:\n1. Prompt the user to input their choice of rock, paper, or scissors.\n2. Generate a random choice for the computer.\n3. Compare the choices and det...
false
wizardlm
wizardlm_0
null
[ { "from": "human", "value": "Enumerate two contrasting dissimilarities between machine learning techniques that are specifically guided by labeled data, known as supervised learning, and the other technique that leverages unstructured data, known as unsupervised learning." }, { "from": "gpt", "v...
false
wizardlm
llavav1.5_580234
ocrvqa/1594161437.jpg
[ { "from": "human", "value": "<image>\nผู้เขียนหนังสือเล่มนี้คือใคร?\nตอบคำถามโดยใช้คำหรือวลีเดียว" }, { "from": "gpt", "value": "จอร์จ ดับเบิลยู. บูเดรอว์" }, { "from": "human", "value": "หนังสือเล่มนี้มีชื่อว่าอะไร?" }, { "from": "gpt", "value": "อิสรภาพ: คู่มือสู่ฟิลาเด...
true
llavav1.5
svit_detail_62659
visual_genome/2371842.jpg
[ { "from": "human", "value": "<image>\nShare an extensive description of the image." }, { "from": "gpt", "value": "The image shows a beautifully presented meal sitting on a wooden tabletop with a marble pattern, likely taken during the day in an indoor setting. A round, white plate sits predomina...
false
svit_detail
wizardlm_0
null
[ { "from": "human", "value": "In addition to the latest population figure for France, please provide the historical trend of population growth or decline over the past 10 years. Furthermore, include the population breakdown by gender and age group in a table format along with any relevant statistical analysi...
false
wizardlm
llavav1.5_184794
coco_2017/000000303047.jpg
[ { "from": "human", "value": "<image>\nWhat is the girl laying on?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Bed" }, { "from": "human", "value": "What is the tattoo on the left ankle?" }, { "from": "gpt", "value": "Stars" }, { ...
false
llavav1.5
llavav1.5_601628
ocrvqa/1592132367.jpg
[ { "from": "human", "value": "<image>\nผู้เขียนหนังสือเล่มนี้คือใคร?\nตอบคำถามโดยใช้คำหรือวลีเดียว" }, { "from": "gpt", "value": "เกิร์ดา เลิร์นเนอร์" }, { "from": "human", "value": "หนังสือเล่มนี้มีชื่อว่าอะไร?" }, { "from": "gpt", "value": "Fireweed: อัตชีวประวัติทางการเ...
true
llavav1.5
llavav1.5_556516
ocrvqa/545215951.jpg
[ { "from": "human", "value": "<image>\nWho is the author of this book?\nAnswer the question using a single word or phrase." }, { "from": "gpt", "value": "Norman Bridwell" }, { "from": "human", "value": "What is the title of this book?" }, { "from": "gpt", "value": "Cliffor...
false
llavav1.5
End of preview.

Typhoon Vision Preview Data

Dataset Overview

This dataset is designed for visual question-answering and image-to-text tasks, supporting both Thai (th) and English (en) languages.

Data Source

The dataset is based on the Bunny image dataset. You can download the original images from here.

Dataset Splits

The dataset is organized into multiple splits, available as Hugging Face datasets:

  1. Pretrain: Used for pretraining the adapter in LLaVA format.
  2. Finetune: Used for finetuning in LLaVA format.
  3. Finetune_translated_stats: Contains original texts, their Thai translations, and COMET scores (translation quality estimation).

Pretraining Set

  • Comprises the original Bunny data.
  • Includes an additional 10% of translated data appended to the original set.
  • COMET QE scores were not computed for this set.

Finetuning Set

  • Based on the same structure as the pretraining set.
  • The appended 10% data consists of top-performing translations, as determined by COMET scores.

File Descriptions

  • pretrain.json: Dataset for pretraining the adapter in LLaVA format.
  • finetune.json: Dataset for finetuning in LLaVA format.
  • finetune_translated_stats.json: Contains original texts, Thai translations, and COMET scores.

Usage Notes

  • The dataset is designed for use with the LLaVA (Large Language and Vision Assistant) format.
  • When using the finetuning set, be aware that it includes high-quality translations based on COMET scores.
  • The pretraining set can be used for initial model adaptation, while the finetuning set is optimized for final model tuning.

Language Support

This dataset supports bilingual tasks:

  • Thai (th)
  • English (en)

Researchers and developers can use this dataset for tasks involving both languages, especially for cross-lingual visual question-answering and image-to-text generation.

Downloads last month
9