Flickr8k
E899058
Flickr8k is a benchmark image-captioning dataset consisting of 8,000 images each paired with multiple human-written descriptions, widely used for training and evaluating vision-language models.
All labels observed (1)
| Label | Occurrences |
|---|---|
| Flickr8k canonical | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T11003511 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Flickr8k Context triple: [Show and Tell: A Neural Image Caption Generator, usesDataset, Flickr8k]
-
A.
Flickr
Flickr is an online photo and video hosting and sharing platform that became one of the earliest popular social media sites for photographers and casual users alike.
-
B.
Britannica ImageQuest
Britannica ImageQuest is a curated educational image database offering millions of rights-cleared photos and illustrations for teaching and learning.
-
C.
Images and Words
Images and Words is a landmark 1992 progressive metal album by Dream Theater, widely credited with bringing the band mainstream recognition and defining their signature sound.
-
D.
IMG
IMG is a global sports, events, and talent management company known for representing athletes and models and producing major sporting and fashion events.
-
E.
Images
Images is a set of impressionistic piano pieces by Claude Debussy that evoke vivid, atmospheric soundscapes through innovative harmony and tone color.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: Flickr8k Target entity description: Flickr8k is a benchmark image-captioning dataset consisting of 8,000 images each paired with multiple human-written descriptions, widely used for training and evaluating vision-language models.
-
A.
Flickr
Flickr is an online photo and video hosting and sharing platform that became one of the earliest popular social media sites for photographers and casual users alike.
-
B.
Britannica ImageQuest
Britannica ImageQuest is a curated educational image database offering millions of rights-cleared photos and illustrations for teaching and learning.
-
C.
Images and Words
Images and Words is a landmark 1992 progressive metal album by Dream Theater, widely credited with bringing the band mainstream recognition and defining their signature sound.
-
D.
IMG
IMG is a global sports, events, and talent management company known for representing athletes and models and producing major sporting and fashion events.
-
E.
Images
Images is a set of impressionistic piano pieces by Claude Debussy that evoke vivid, atmospheric soundscapes through innovative harmony and tone color.
- F. None of above. chosen
Statements (46)
| Predicate | Object |
|---|---|
| instanceOf |
benchmark dataset
ⓘ
image-captioning dataset ⓘ vision-language dataset ⓘ |
| hasAnnotationType |
image captions
ⓘ
natural language descriptions ⓘ |
| hasApproximateNumberOfImages | 8000 ⓘ |
| hasBenchmarkRole | baseline dataset for image captioning ⓘ |
| hasCaptionSource | human annotators ⓘ |
| hasCaptionsPerImage | 5 ⓘ |
| hasDataModality |
images
ⓘ
text captions ⓘ |
| hasDataType | photographic images ⓘ |
| hasDescriptionQuality | human-written captions ⓘ |
| hasDomain |
computer vision
ⓘ
natural language processing ⓘ |
| hasEvaluationMetrics |
BLEU
NERFINISHED
ⓘ
CIDEr NERFINISHED ⓘ METEOR NERFINISHED ⓘ ROUGE NERFINISHED ⓘ |
| hasImageSource | Flickr NERFINISHED ⓘ |
| hasLanguage | English ⓘ |
| hasLicenseType | research use ⓘ |
| hasName | Flickr8k NERFINISHED ⓘ |
| hasNumberOfImages | 8000 ⓘ |
| hasScale | small-scale image-captioning dataset ⓘ |
| hasSource | Flickr NERFINISHED ⓘ |
| hasTask |
multimodal learning
ⓘ
vision-language grounding ⓘ |
| hasTypicalSplit |
test set
ⓘ
training set ⓘ validation set ⓘ |
| isBenchmarkFor |
automatic image description
ⓘ
multimodal representation learning ⓘ |
| isConsidered | standard benchmark in image captioning ⓘ |
| isSmallerThan |
Flickr30k
NERFINISHED
ⓘ
MS COCO NERFINISHED ⓘ |
| isUsedIn | vision-and-language research ⓘ |
| isUsedTo |
compare image-captioning algorithms
ⓘ
evaluate caption generation quality ⓘ |
| isWidelyUsedFor | benchmarking captioning models ⓘ |
| isWidelyUsedIn | academic research ⓘ |
| usedFor |
evaluating image-captioning models
ⓘ
evaluating vision-language models ⓘ image captioning ⓘ training image-captioning models ⓘ training vision-language models ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: Flickr8k Description of subject: Flickr8k is a benchmark image-captioning dataset consisting of 8,000 images each paired with multiple human-written descriptions, widely used for training and evaluating vision-language models.
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.