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.
Language Knowledge-Assisted Representation Learning for Skeleton-Based Action Recognition
SCP: Soft Conditional Prompt Learning for Aerial Video Action Recognition
GPT-3.5, GPT-4, or BARD? Evaluating LLMs Reasoning Ability in Zero-Shot Setting and Performance Boosting Through Prompts
CoMusion: Towards Consistent Stochastic Human Motion Prediction via Motion Diffusion
Do Models Really Learn to Follow Instructions? An Empirical Study of Instruction Tuning
Remembering What Is Important: A Factorised Multi-Head Retrieval and Auxiliary Memory Stabilisation Scheme for Human Motion Prediction
Overcoming Topology Agnosticism: Enhancing Skeleton-Based Action Recognition through Redefined Skeletal Topology Awareness
Introspective Tips: Large Language Model for In-Context Decision Making
LLM-Pruner: On the Structural Pruning of Large Language Models
Pengi: An Audio Language Model for Audio Tasks
Boosting Crop Classification by Hierarchically Fusing Satellite, Rotational, and Contextual Data
ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval
MedBLIP: Bootstrapping Language-Image Pre-training from 3D Medical Images and Texts
Riemannian Multinomial Logistics Regression for SPD Neural Networks
Motion-Scenario Decoupling for Rat-Aware Video Position Prediction: Strategy and Benchmark
TSGN: Temporal Scene Graph Neural Networks with Projected Vectorized Representation for Multi-Agent Motion Prediction
ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding
Dual Intent Enhanced Graph Neural Network for Session-based New Item Recommendation
Few-shot Action Recognition via Intra- and Inter-Video Information Maximization
Learning Video-Conditioned Policies for Unseen Manipulation Tasks
Alternating Gradient Descent and Mixture-of-Experts for Integrated Multimodal Perception
Boosting Visual-Language Models by Exploiting Hard Samples
ImageBind: One Embedding Space To Bind Them All
Unlocking Practical Applications in Legal Domain: Evaluation of GPT for Zero-Shot Semantic Annotation of Legal Texts
Towards Accurate Human Motion Prediction via Iterative Refinement
Building Neural Networks on Matrix Manifolds: A Gyrovector Space Approach
Video-Specific Query-Key Attention Modeling for Weakly-Supervised Temporal Action Localization
Analyzing Hong Kong's Legal Judgments from a Computational Linguistics point-of-view
LLM2Loss: Leveraging Language Models for Explainable Model Diagnostics
Unsupervised Improvement of Audio-Text Cross-Modal Representations
The Benefits of Label-Description Training for Zero-Shot Text Classification
Cross-Stream Contrastive Learning for Self-Supervised Skeleton-Based Action Recognition
Self-similarity-based super-resolution of photoacoustic angiography from hand-drawn doodles
DRPT: Disentangled and Recurrent Prompt Tuning for Compositional Zero-Shot Learning
Stance Detection: A Practical Guide to Classifying Political Beliefs in Text
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.