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.
Modeling Caption Diversity in Contrastive Vision-Language Pretraining
Enhancing Action Recognition from Low-Quality Skeleton Data via Part-Level Knowledge Distillation
Prevalent Frequency of Emotional and Physical Symptoms in Social Anxiety using Zero Shot Classification: An Observational Study
Dual Expert Distillation Network for Generalized Zero-Shot Learning
An Improved Graph Pooling Network for Skeleton-Based Action Recognition
OpenDlign: Open-World Point Cloud Understanding with Depth-Aligned Images
Zero-Shot Distillation for Image Encoders: How to Make Effective Use of Synthetic Data
Embracing Diversity: Interpretable Zero-shot classification beyond one vector per class
AAPL: Adding Attributes to Prompt Learning for Vision-Language Models
Motor Focus: Fast Ego-Motion Prediction for Assistive Visual Navigation
HDBN: A Novel Hybrid Dual-branch Network for Robust Skeleton-based Action Recognition
Visual-Augmented Dynamic Semantic Prototype for Generative Zero-Shot Learning
TAXI: Evaluating Categorical Knowledge Editing for Language Models
Adversarial Representation Engineering: A General Model Editing Framework for Large Language Models
STAT: Towards Generalizable Temporal Action Localization
FipTR: A Simple yet Effective Transformer Framework for Future Instance Prediction in Autonomous Driving
PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation
DeepLocalization: Using change point detection for Temporal Action Localization
Small Language Models are Good Too: An Empirical Study of Zero-Shot Classification
Decomposing and Editing Predictions by Modeling Model Computation
State-space Decomposition Model for Video Prediction Considering Long-term Motion Trend
MemLLM: Finetuning LLMs to Use An Explicit Read-Write Memory
MK-SGN: A Spiking Graph Convolutional Network with Multimodal Fusion and Knowledge Distillation for Skeleton-based Action Recognition
Two-Stage Stance Labeling: User-Hashtag Heuristics with Graph Neural Networks
Learning to Score Sign Language with Two-stage Method
Leveraging Temporal Contextualization for Video Action Recognition
CREST: Cross-modal Resonance through Evidential Deep Learning for Enhanced Zero-Shot Learning
The Devil is in the Few Shots: Iterative Visual Knowledge Completion for Few-shot Learning
Evolving Interpretable Visual Classifiers with Large Language Models
Knowledge-enhanced Visual-Language Pretraining for Computational Pathology
Black-box Adversarial Transferability: An Empirical Study in Cybersecurity Perspective
In My Perspective, In My Hands: Accurate Egocentric 2D Hand Pose and Action Recognition
A Survey on Multimodal Wearable Sensor-based Human Action Recognition
Customising General Large Language Models for Specialised Emotion Recognition Tasks
Exploring Explainability in Video Action Recognition
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.