Zero-Shot Learning
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 7
- 평가표
- 33
- 보존된 평가 기록
- 84
Zero-shot learning (ZSL) is a model's ability to detect classes never seen during training. The condition is that the classes are not known during supervised learning.
Earlier work in zero-shot learning use attributes in a two-step approach to infer unknown classes. In the computer vision context, more recent advances learn mappings from image feature space to semantic space. Other approaches learn non-linear multimodal embeddings. In the modern NLP context, language models can be evaluated on downstream tasks without fine tuning.
Benchmark datasets for zero-shot learning include aPY, AwA, and CUB, among others.
( Image credit: Prototypical Networks for Few shot Learning in PyTorch )
Further readings:
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
A Closer Look at the Explainability of Contrastive Language-Image Pre-training
Best Practices for 2-Body Pose Forecasting
RECLIP: Resource-efficient CLIP by Training with Small Images
FollowMe: Vehicle Behaviour Prediction in Autonomous Vehicle Settings
WEAR: An Outdoor Sports Dataset for Wearable and Egocentric Activity Recognition
Evaluation of Differentially Constrained Motion Models for Graph-Based Trajectory Prediction
DeFeeNet: Consecutive 3D Human Motion Prediction with Deviation Feedback
Interaction-Aware Prompting for Zero-Shot Spatio-Temporal Action Detection
SAM.MD: Zero-shot medical image segmentation capabilities of the Segment Anything Model
Attack-Augmentation Mixing-Contrastive Skeletal Representation Learning
Graph-Guided MLP-Mixer for Skeleton-Based Human Motion Prediction
Synthetic Sample Selection for Generalized Zero-Shot Learning
TM2D: Bimodality Driven 3D Dance Generation via Music-Text Integration
Structured prompt interrogation and recursive extraction of semantics (SPIRES): A method for populating knowledge bases using zero-shot learning
Zero-shot Medical Image Translation via Frequency-Guided Diffusion Models
A Deep Multi-Modal Cyber-Attack Detection in Industrial Control Systems
Exploring Vision-Language Models for Imbalanced Learning
A real-time algorithm for human action recognition in RGB and thermal video
Learning to Name Classes for Vision and Language Models
Focalized Contrastive View-invariant Learning for Self-supervised Skeleton-based Action Recognition
AutoLabel: CLIP-based framework for Open-set Video Domain Adaptation
On the Benefits of 3D Pose and Tracking for Human Action Recognition
HaLP: Hallucinating Latent Positives for Skeleton-based Self-Supervised Learning of Actions
Semi-Weakly Supervised Object Kinematic Motion Prediction
STMT: A Spatial-Temporal Mesh Transformer for MoCap-Based Action Recognition
FIR-based Future Trajectory Prediction in Nighttime Autonomous Driving
HARFLOW3D: A Latency-Oriented 3D-CNN Accelerator Toolflow for HAR on FPGA Devices
JCDNet: Joint of Common and Definite phases Network for Weakly Supervised Temporal Action Localization
SoftCLIP: Softer Cross-modal Alignment Makes CLIP Stronger
VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking
Zero-shot Entailment of Leaderboards for Empirical AI Research
Evaluation of ChatGPT for NLP-based Mental Health Applications
Variational Distribution Learning for Unsupervised Text-to-Image Generation
Your Diffusion Model is Secretly a Zero-Shot Classifier
Structured Video-Language Modeling with Temporal Grouping and Spatial Grounding
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.