https://github.com/escorciav/Mixture-of-Embedding-Experts
Learning a Text-Video Embedding from Incomplete and Heterogeneous Data
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
A methodology that involves selecting relevant data or examples from a large dataset to support tasks like prediction, learning, or inference. It enhances models by providing context or additional information, often used in systems like retrieval-augmented generation or in-context learning.
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Learning a Text-Video Embedding from Incomplete and Heterogeneous Data
Learning to Cluster Faces via Confidence and Connectivity Estimation
SegMap: Segment-based mapping and localization using data-driven descriptors
ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis
COCO-CN for Cross-Lingual Image Tagging, Captioning and Retrieval
VLGrammar: Grounded Grammar Induction of Vision and Language
ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT
RAGAS: Automated Evaluation of Retrieval Augmented Generation
CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval
Improving zero-shot learning by mitigating the hubness problem
EDIN: An End-to-end Benchmark and Pipeline for Unknown Entity Discovery and Indexing
VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding
Loss in Translation: Learning Bilingual Word Mapping with a Retrieval Criterion
Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation
Circulant temporal encoding for video retrieval and temporal alignment
Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination
DeepDiary: Automatic Caption Generation for Lifelogging Image Streams
PatentMatch: A Dataset for Matching Patent Claims & Prior Art
Nearly Minimax-Optimal Rates for Noisy Sparse Phase Retrieval via Early-Stopped Mirror Descent
Neural Collaborative Filtering vs. Matrix Factorization Revisited
DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.