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.
OFMPNet: Deep End-to-End Model for Occupancy and Flow Prediction in Urban Environment
Diffusion based Zero-shot Medical Image-to-Image Translation for Cross Modality Segmentation
LoSA: Long-Short-range Adapter for Scaling End-to-End Temporal Action Localization
LLMs are Good Action Recognizers
Dual DETRs for Multi-Label Temporal Action Detection
Training-Free Semantic Segmentation via LLM-Supervision
An Unsupervised Adversarial Autoencoder for Cyber Attack Detection in Power Distribution Grids
Is Factuality Enhancement a Free Lunch For LLMs? Better Factuality Can Lead to Worse Context-Faithfulness
Small Language Models Learn Enhanced Reasoning Skills from Medical Textbooks
Dialectical Alignment: Resolving the Tension of 3H and Security Threats of LLMs
Multi-hop Question Answering under Temporal Knowledge Editing
Retrieval-enhanced Knowledge Editing in Language Models for Multi-Hop Question Answering
X-MIC: Cross-Modal Instance Conditioning for Egocentric Action Generalization
LC-LLM: Explainable Lane-Change Intention and Trajectory Predictions with Large Language Models
An Efficient Risk-aware Branch MPC for Automated Driving that is Robust to Uncertain Vehicle Behaviors
PLOT-TAL -- Prompt Learning with Optimal Transport for Few-Shot Temporal Action Localization
EgoNav: Egocentric Scene-aware Human Trajectory Prediction
Robust and Scalable Model Editing for Large Language Models
CMP: Cooperative Motion Prediction with Multi-Agent Communication
ProIn: Learning to Predict Trajectory Based on Progressive Interactions for Autonomous Driving
VP3D: Unleashing 2D Visual Prompt for Text-to-3D Generation
DeGCN: Deformable Graph Convolutional Networks for Skeleton-Based Action Recognition
Gaze-guided Hand-Object Interaction Synthesis: Dataset and Method
General LLMs as Instructors for Domain-Specific LLMs: A Sequential Fusion Method to Integrate Extraction and Editing
VLM-CPL: Consensus Pseudo Labels from Vision-Language Models for Human Annotation-Free Pathological Image Classification
Human Motion Prediction under Unexpected Perturbation
Comprehensive Evaluation and Insights into the Use of Large Language Models in the Automation of Behavior-Driven Development Acceptance Test Formulation
GCN-DevLSTM: Path Development for Skeleton-Based Action Recognition
InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
Long-CLIP: Unlocking the Long-Text Capability of CLIP
Existence Is Chaos: Enhancing 3D Human Motion Prediction with Uncertainty Consideration
A Unified Framework for Model Editing
Less but Better: Enabling Generalized Zero-shot Learning Towards Unseen Domains by Intrinsic Learning from Redundant LLM Semantics
Editing Knowledge Representation of Language Model via Rephrased Prefix Prompts
Locating and Mitigating Gender Bias in Large Language Models
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.