image-sentence alignment
12개 벤치마크 · 논문 3편 · 이 태스크의 논문 보기 →
Benchmarks
VALSE
결과 14개
VALSE actant swap
결과 14개
VALSE action replacement
결과 14개
VALSE coreference clean
결과 14개
VALSE coreference standard
결과 14개
VALSE counting adversarial
결과 14개
VALSE counting balanced
결과 14개
VALSE counting small numbers
결과 14개
VALSE existence
결과 14개
VALSE foil-it (noun phrases)
결과 14개
VALSE plurality
결과 14개
VALSE spatial relations
결과 14개
Most implemented
Papers
VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena
2021-12-14 · ACL 2022 5
· Letitia Parcalabescu, Michele Cafagna, Lilitta Muradjan, Anette Frank 외
We propose VALSE (Vision And Language Structured Evaluation), a novel benchmark designed for testing general-purpose pretrained vision and language (V&L) models for their visio-linguistic grounding capabilities on specif…
image-sentence alignmentvalidRemoving Word-Level Spurious Alignment between Images and Pseudo-Captions in Unsupervised Image Captioning
2021-04-28 · EACL 2021 2
· Ukyo Honda, Yoshitaka Ushiku, Atsushi Hashimoto, Taro Watanabe 외
Unsupervised image captioning is a challenging task that aims at generating captions without the supervision of image-sentence pairs, but only with images and sentences drawn from different sources and object labels dete…
Image Captioningimage-sentence alignmentObjectSentenceFrom Images to Sentences through Scene Description Graphs using Commonsense Reasoning and Knowledge
2015-11-10
· Somak Aditya, Yezhou Yang, Chitta Baral, Cornelia Fermuller 외
In this paper we propose the construction of linguistic descriptions of images. This is achieved through the extraction of scene description graphs (SDGs) from visual scenes using an automatically constructed knowledge b…
image-sentence alignmentSentence