HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Image Text Matching On Commercialadsdataset

평가 지표

ADD(S) AUC

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	ADD(S) AUC	Paper Title	Repository
OSCAR	87.45	Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks
Unicoder-VL	83.16	Unicoder-VL: A Universal Encoder for Vision and Language by Cross-modal Pre-training	-
VinVL	88.56	VinVL: Revisiting Visual Representations in Vision-Language Models
BLIP	83.51	BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
AlignCMSS	91.73	Align before Search: Aligning Ads Image to Text for Accurate Cross-Modal Sponsored Search
ALBEF	82.74	Align before Fuse: Vision and Language Representation Learning with Momentum Distillation
VL-BERT	86.27	VL-BERT: Pre-training of Generic Visual-Linguistic Representations
AdsCVLR	87.90	AdsCVLR: Commercial Visual-Linguistic Representation Modeling in Sponsored Search	-

0 of 8 row(s) selected.