paper-with-me

홈 › Papers

Efficient Discriminative Joint Encoders for Large Scale Vision-Language Reranking

2025-10-08 · Mitchell Keren Taraday, Shahaf Wagner, Chaim Baskin arxiv

Multimodal retrieval still leans on embedding-based models like CLIP for fast vector search over pre-computed image embeddings. Yet, unlike text retrieval, where joint-encoder rerankers are standard, comparable vision-language rerankers are largely absent. We find that seminal joint encoders such as BLIP are severely bottlenecked by an expensive visual feature-extraction stage, preventing practical deployment at scale. Motivated by this bottleneck, we introduce EDJE, an Efficient Discriminative Joint Encoder that precomputes vision tokens offline and compresses them via a lightweight attention-based adapter, so online inference runs only a compact joint encoder over a small set of visual tokens plus the text. EDJE preserves strong retrieval performance while drastically reducing storage and online compute, enabling high-throughput inference. Specifically, EDJE processes 50k image--text pairs/second while requiring 49kB of disk storage per image, matching prior art on Flickr (zero-shot) and COCO (fine-tuned) retrieval.

📄 PDF Abstract BibTeX arXiv:2510.06820

Code (0)

등록된 구현이 없습니다.

Tasks

Text Retrieval

Similar Papers 제목 키워드 기반

Discriminatively Boosted Image Clustering with Fully Convolutional Auto-Encoders

2017-03-23 · Fengfu Li, Hong Qiao, Bo Zhang, Xuanyang Xi

Traditional image clustering methods take a two-step approach, feature learning and clustering, sequentially. However, recent research results demonstrated that combining the separated phases in a unified framework and t…

ClusteringImage Clustering

Denoising Diffusion Autoencoders are Unified Self-supervised Learners

2023-03-17 · ICCV 2023 1 · Weilai Xiang, Hongyu Yang, Di Huang, Yunhong Wang

Inspired by recent advances in diffusion models, which are reminiscent of denoising autoencoders, we investigate whether they can acquire discriminative representations for classification via generative pre-training. Thi…

Contrastive LearningDenoisingImage GenerationLinear evaluation+2

Explicit Time-Frequency Dynamics for Skeleton-Based Gait Recognition

2026-04-03 · Seoyeon Ko, Yeojin Song, Egene Chung, Luca Quagliato 외 arxiv

Skeleton-based gait recognizers excel at modeling spatial configurations but often underuse explicit motion dynamics that are crucial under appearance changes. We introduce a plug-and-play Wavelet Feature Stream that aug…

Gait Recognition

Unlocking Few-Shot Capabilities in LVLMs via Prompt Conditioning and Head Selection

2026-03-25 · Adhemar de Senneville, Xavier Bou, Jérémy Anger, Rafael Grompone 외 arxiv

Current Large Vision Language Models (LVLMs) excel at many zero-shot tasks like image captioning, visual question answering and OCR. However, these same models suffer from poor performance at image classification tasks, …

Visual Question AnsweringImage ClassificationImage Captioning

VladVA: Discriminative Fine-tuning of LVLMs

2024-12-05 · CVPR 2025 1 · Yassine Ouali, Adrian Bulat, Alexandros Xenos, Anestis Zaganidis 외

Contrastively-trained Vision-Language Models (VLMs) like CLIP have become the de facto approach for discriminative vision-language representation learning. However, these models have limited language understanding, often…

Image-text RetrievalRepresentation LearningText Retrieval