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.
Enabling Small Models for Zero-Shot Selection and Reuse through Model Label Learning
XDT-CXR: Investigating Cross-Disease Transferability in Zero-Shot Binary Classification of Chest X-Rays
Memorization in In-Context Learning
Exploring Large Language Models for Feature Selection: A Data-centric Perspective
A Review of Human-Object Interaction Detection
Beneath the Surface of Consistency: Exploring Cross-lingual Knowledge Representation Sharing in LLMs
TDS-CLIP: Temporal Difference Side Network for Image-to-Video Transfer Learning
AnyGraph: Graph Foundation Model in the Wild
MEGen: Generative Backdoor in Large Language Models via Model Editing
CrossFi: A Cross Domain Wi-Fi Sensing Framework Based on Siamese Network
Event Stream based Human Action Recognition: A High-Definition Benchmark Dataset and Algorithms
Cross-composition Feature Disentanglement for Compositional Zero-shot Learning
Attribution Analysis Meets Model Editing: Advancing Knowledge Correction in Vision Language Models with VisEdit
SHARP: Segmentation of Hands and Arms by Range using Pseudo-Depth for Enhanced Egocentric 3D Hand Pose Estimation and Action Recognition
Resolving Lexical Bias in Edit Scoping with Projector Editor Networks
CLIP-DPO: Vision-Language Models as a Source of Preference for Fixing Hallucinations in LVLMs
ELDER: Enhancing Lifelong Model Editing with Mixture-of-LoRA
Joint Temporal Pooling for Improving Skeleton-based Action Recognition
CLIP-CID: Efficient CLIP Distillation via Cluster-Instance Discrimination
MambaTrack: A Simple Baseline for Multiple Object Tracking with State Space Model
Flatten: Video Action Recognition is an Image Classification task
TsCA: On the Semantic Consistency Alignment via Conditional Transport for Compositional Zero-Shot Learning
EasyRec: Simple yet Effective Language Models for Recommendation
Navigating Data Scarcity using Foundation Models: A Benchmark of Few-Shot and Zero-Shot Learning Approaches in Medical Imaging
Towards Practical Human Motion Prediction with LiDAR Point Clouds
Zero-Shot Learning and Key Points Are All You Need for Automated Fact-Checking
Action Recognition for Privacy-Preserving Ambient Assisted Living
Knowledge in Superposition: Unveiling the Failures of Lifelong Knowledge Editing for Large Language Models
DC3DO: Diffusion Classifier for 3D Objects
Probabilistic Vision-Language Representation for Weakly Supervised Temporal Action Localization
OmniCLIP: Adapting CLIP for Video Recognition with Spatial-Temporal Omni-Scale Feature Learning
HAT: History-Augmented Anchor Transformer for Online Temporal Action Localization
Attention Please: What Transformer Models Really Learn for Process Prediction
Unleashing The Power of Pre-Trained Language Models for Irregularly Sampled Time Series
A Training-Free Framework for Video License Plate Tracking and Recognition with Only One-Shot
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.