PiTL: Cross-modal Retrieval with Weakly-supervised Vision-language Pre-training via Prompting

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

41 Lataukset (Pure)

Abstrakti

Vision-language (VL) Pre-training (VLP) has shown to well generalize VL models over a wide range of VL downstream tasks, especially for cross-modal retrieval. However, it hinges on a huge amount of image-text pairs, which requires tedious and costly curation. On the contrary, weakly-supervised VLP (W-VLP) [33] explores means with object tags generated by a pre-trained object detector (OD) from images. Yet, they still require paired information, i.e. images and object-level annotations, as supervision to train an OD. To further reduce the amount of supervision, we propose Prompts-in-The-Loop (PiTL) that prompts knowledge from large language models (LLMs) to describe images. Concretely, given a category label of an image, e.g. refinery, the knowledge, e.g. a refinery could be seen with large storage tanks, pipework, and..., extracted by LLMs is used as the language counterpart. The knowledge supplements, e.g. the common relations among entities most likely appearing in a scene. We create IN14K, a new VL dataset of 9M images and 1M descriptions of 14K categories from ImageNet21K [8] with PiTL. Empirically, the VL models pre-trained with PiTL-generated pairs are strongly favored over other W-VLP works on image-to-text (I2T) and text-to-image (T2I) retrieval tasks, with less supervision. The results reveal the effectiveness of PiTL-generated pairs for VLP.

AlkuperäiskieliEnglanti
OtsikkoSIGIR 2023 - Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval
KustantajaACM
Sivut2261-2265
Sivumäärä5
ISBN (elektroninen)978-1-4503-9408-6
DOI - pysyväislinkit
TilaJulkaistu - 19 heinäk. 2023
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaInternational ACM SIGIR Conference on Research and Development in Information Retrieval - Taipei, Taiwan
Kesto: 23 heinäk. 202327 heinäk. 2023
Konferenssinumero: 46

Conference

ConferenceInternational ACM SIGIR Conference on Research and Development in Information Retrieval
LyhennettäSIGIR
Maa/AlueTaiwan
KaupunkiTaipei
Ajanjakso23/07/202327/07/2023

Sormenjälki

Sukella tutkimusaiheisiin 'PiTL: Cross-modal Retrieval with Weakly-supervised Vision-language Pre-training via Prompting'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä