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.
Label Set Optimization via Activation Distribution Kurtosis for Zero-shot Classification with Generative Models
Inference time LLM alignment in single and multidomain preference spectrum
Unsupervised Domain Adaptation for Action Recognition via Self-Ensembling and Conditional Embedding Alignment
Backdoor in Seconds: Unlocking Vulnerabilities in Large Pre-trained Models via Model Editing
SpikMamba: When SNN meets Mamba in Event-based Human Action Recognition
Pedestrian motion prediction evaluation for urban autonomous driving
Representation Shattering in Transformers: A Synthetic Study with Knowledge Editing
Gradient Rewiring for Editable Graph Neural Network Training
LiMTR: Time Series Motion Prediction for Diverse Road Users through Multimodal Feature Integration
Improving the Multi-label Atomic Activity Recognition by Robust Visual Feature and Advanced Attention @ ROAD++ Atomic Activity Recognition 2024
MoRE: Multi-Modal Contrastive Pre-training with Transformers on X-Rays, ECGs, and Diagnostic Report
Can Knowledge Editing Really Correct Hallucinations?
Assessing Open-world Forgetting in Generative Image Model Customization
Storyboard guided Alignment for Fine-grained Video Action Recognition
EcomEdit: An Automated E-commerce Knowledge Editing Framework for Enhanced Product and Purchase Intention Understanding
Can Medical Vision-Language Pre-training Succeed with Purely Synthetic Data?
Can We Reverse In-Context Knowledge Edits?
Towards Graph Foundation Models: Training on Knowledge Graphs Enables Transferability to General Graphs
Mechanistic Unlearning: Robust Knowledge Unlearning and Editing via Mechanistic Localization
LLM Chain Ensembles for Scalable and Accurate Data Annotation
Interpreting and Analysing CLIP's Zero-Shot Image Classification via Mutual Knowledge
MF-LAL: Drug Compound Generation Using Multi-Fidelity Latent Space Active Learning
O-Edit: Orthogonal Subspace Editing for Language Model Sequential Editing
PSVMA+: Exploring Multi-granularity Semantic-visual Adaption for Generalized Zero-shot Learning
CtrlSynth: Controllable Image Text Synthesis for Data-Efficient Multimodal Learning
Continual Learning Improves Zero-Shot Action Recognition
LoRD: Adapting Differentiable Driving Policies to Distribution Shifts
Multi class activity classification in videos using Motion History Image generation
Retrieval Instead of Fine-tuning: A Retrieval-based Parameter Ensemble for Zero-shot Learning
Large Model for Small Data: Foundation Model for Cross-Modal RF Human Activity Recognition
Real-time Monitoring of Lower Limb Movement Resistance Based on Deep Learning
Keys to Robust Edits: from Theoretical Insights to Practical Advances
ESVO2: Direct Visual-Inertial Odometry with Stereo Event Cameras
CollabEdit: Towards Non-destructive Collaborative Knowledge Editing
SmartPretrain: Model-Agnostic and Dataset-Agnostic Representation Learning for Motion Prediction
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.