CVPR2023论文整理

---

CVPR2023

根据官方信息统计，今年共收到 9155 份提交，比去年增加了 12%，创下新纪录，今年接收了 2360 篇论文，接收率为 25.78%。作为对比，去年有 8100 多篇有效投稿，大会接收了 2067 篇，接收率为 25%。

https://cvpr2023.thecvf.com/Conferences/2023/AcceptedPapers

现在根据关键词，对自己感兴趣的方向进行规整以及分类（有筛选）

一. Vision and Language / Multimodal

论文名	简介
Improving Commonsense in Vision-Language Models via Knowledge Graph Riddles
Filtering, Distillation, and Hard Negatives for Vision-Language Pre-Training
Seeing What You Miss: Vision-Language Pre-training with Semantic Completion Learning
Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks
CREPE: Can Vision-Language Foundation Models Reason Compositionally?
Task Residual for Tuning Vision-Language Models
Q: How to Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A Self-Train on Unlabeled Images!
FAME-ViL: Multi-Tasking Vision-Language Model for Heterogeneous Fashion Tasks
VILA: Learning Image Aesthetics from User Comments with Vision-Language Pretraining
Open-set Fine-grained Retrieval via Prompting Vision-Language Evaluator
Image as a Foreign Language BEiT Pretraining for Vision and Vision-Language Tasks
FashionSAP: Symbols and Attributes Prompt for Fine-grained Fashion Vision-Language Pre-training
Accelerating Vision-Language Pretraining with Free Language Modeling
Leveraging per Image-Token Consistency for Vision-Language Pre-training
Position-guided Text Prompt for Vision-Language Pre-training
IFSeg: Image-free Semantic Segmentation via Vision-Language Model
Enhanced Multimodal Representation Learning with Cross-modal KD
Efficient Multimodal Fusion via Interactive Prompting
Best of Both Worlds: Multimodal Contrastive Learning with Tabular and Imaging Data
Revisiting Multimodal Representation in Contrastive Learning From Patch and Token embeddings to Finite Discrete Tokens
Align and Attend: Multimodal Summarization with Dual Contrastive Losses
Multimodal Prompting with Missing Modalities for Visual Recognition