SOTAVerified

Zero-Shot Learning

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:

Papers

Showing 9511000 of 1864 papers

TitleStatusHype
Simple Primitives with Feasibility- and Contextuality-Dependence for Open-World Compositional Zero-shot Learning0
Crosslingual Generalization through Multitask FinetuningCode2
Fine-grained Visual-Text Prompt-Driven Self-Training for Open-Vocabulary Object Detection0
Chinese CLIP: Contrastive Vision-Language Pretraining in ChineseCode5
End-to-end Spoken Language Understanding with Tree-constrained Pointer GeneratorCode0
Beyond Prompting: Making Pre-trained Language Models Better Zero-shot Learners by Clustering RepresentationsCode1
Being Comes from Not-being: Open-vocabulary Text-to-Motion Generation with Wordless TrainingCode1
Towards Reliable Zero Shot Classification in Self-Supervised Models with Conformal Prediction0
Text2Model: Text-based Model Induction for Zero-shot Image Classification0
Don't Prompt, Search! Mining-based Zero-Shot Learning with Language Models0
Zero-Shot Learning of a Conditional Generative Adversarial Network for Data-Free Network Quantization0
TAPE: Assessing Few-shot Russian Language UnderstandingCode0
ProGen: Progressive Zero-shot Dataset Generation via In-context FeedbackCode1
Learning Attention Propagation for Compositional Zero-Shot Learning0
General Image Descriptors for Open World Image Retrieval using ViT CLIPCode1
TabLLM: Few-shot Classification of Tabular Data with Large Language ModelsCode2
Meta-Learning via Classifier(-free) Diffusion GuidanceCode1
Improving Object-centric Learning with Query OptimizationCode1
Non-Contrastive Learning Meets Language-Image Pre-TrainingCode0
Zero-Shot Learners for Natural Language Understanding via a Unified Multiple Choice PerspectiveCode4
LAION-5B: An open large-scale dataset for training next generation image-text modelsCode0
Rebalanced Zero-shot LearningCode0
Towards a Unified Multi-Dimensional Evaluator for Text GenerationCode2
Visual Classification via Description from Large Language ModelsCode1
Efficient Gaussian Process Model on Class-Imbalanced Datasets for Generalized Zero-Shot Learning0
Contrastive Training Improves Zero-Shot Classification of Semi-structured Documents0
Non-intrusive Load Monitoring based on Self-supervised Learning0
APE: Aligning Pretrained Encoders to Quickly Learn Aligned Multimodal Representations0
LOCL: Learning Object-Attribute Composition using LocalizationCode0
Self-omics: A Self-supervised Learning Framework for Multi-omics Cancer DataCode0
LASP: Text-to-Text Optimization for Language-Aware Soft Prompting of Vision & Language ModelsCode1
ITAINNOVA at SocialDisNER: A Transformers cocktail for disease identification in social media in Spanish0
Zero-shot Script Parsing0
Slot Dependency Modeling for Zero-Shot Cross-Domain Dialogue State Tracking0
HyperHawkes: Hypernetwork based Neural Temporal Point Process0
Prompt-guided Scene Generation for 3D Zero-Shot Learning0
CALIP: Zero-Shot Enhancement of CLIP with Parameter-free AttentionCode1
Lex2Sent: A bagging approach to unsupervised sentiment analysisCode0
DeepGraphONet: A Deep Graph Operator Network to Learn and Zero-shot Transfer the Dynamic Response of Networked Systems0
I2DFormer: Learning Image to Document Attention for Zero-Shot Image Classification0
WeLM: A Well-Read Pre-trained Language Model for Chinese0
Natural Language Inference Prompts for Zero-shot Emotion Classification in Text across CorporaCode1
Vision Transformers for Action Recognition: A Survey0
VL-Taboo: An Analysis of Attribute-based Zero-shot Capabilities of Vision-Language ModelsCode0
FETA: Towards Specializing Foundation Models for Expert Task ApplicationsCode1
Depression Symptoms Modelling from Social Media Text: A Semi-supervised Learning Approach0
Federated Zero-Shot Learning for Visual Recognition0
Design of the topology for contrastive visual-textual alignmentCode0
Flexible Job Classification with Zero-Shot Learning0
Exploring Semantic Attributes from A Foundation Model for Federated Learning of Disjoint Label Spaces0
Show:102550
← PrevPage 20 of 38Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ZeroDiffaverage top-1 classification accuracy87.5Unverified
2DUETaverage top-1 classification accuracy72.3Unverified
3Composeraverage top-1 classification accuracy69.4Unverified
4HDC-ZSC-MLPaverage top-1 classification accuracy65.6Unverified
5ZSL_TF-VAEGANaverage top-1 classification accuracy64.9Unverified
6ZLaPAccuracy64.3Unverified
7ZLaP*Accuracy64.2Unverified
8HDC-ZSCaverage top-1 classification accuracy63.8Unverified
9SPOTaverage top-1 classification accuracy62.9Unverified
10f-VAEGAN-D2average top-1 classification accuracy61Unverified
#ModelMetricClaimedVerifiedStatus
1dmis-lab/biobert-v1.1Accuracy26.15Unverified
2meta-llama/Meta-Llama-3-8B-InstructAccuracy25.84Unverified
3epfl-llm/meditron-7bAccuracy25.75Unverified
4dmis-lab/meerkat-7b-v1.0Accuracy25.68Unverified
5meta-llama/Meta-Llama-3-8B-InstructAccuracy25.65Unverified
6HuggingFaceH4/zephyr-7b-betaAccuracy25.54Unverified
7dmis-lab/biobert-v1.1Accuracy25.46Unverified
8epfl-llm/meditron-70bAccuracy25.36Unverified
9epfl-llm/meditron-70bAccuracy25.26Unverified
10HuggingFaceH4/zephyr-7b-betaAccuracy25.06Unverified
#ModelMetricClaimedVerifiedStatus
1ZeroDiffaverage top-1 classification accuracy77.3Unverified
2SPOT (VAEGAN)average top-1 classification accuracy66.04Unverified
3ZSL_TF-VAEGANaverage top-1 classification accuracy66Unverified
4f-VAEGANaverage top-1 classification accuracy64.7Unverified
5DUET (Ours)average top-1 classification accuracy64.4Unverified
6LisGANaverage top-1 classification accuracy61.7Unverified
7TCNaverage top-1 classification accuracy61.5Unverified
8f-CLSWGANaverage top-1 classification accuracy60.8Unverified
9Cycle-WGANaverage top-1 classification accuracy59.9Unverified
#ModelMetricClaimedVerifiedStatus
1ZeroDiffaverage top-1 classification accuracy86.4Unverified
2ZSL-KGaverage top-1 classification accuracy78.08Unverified
3ZSL_TF-VAEGANaverage top-1 classification accuracy72.2Unverified
4DUET (Ours)average top-1 classification accuracy69.9Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy84Unverified
2ZLaP*Accuracy83.1Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy93.6Unverified
2ZLaPAccuracy93.4Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy74.2Unverified
2ZLaPAccuracy74Unverified
#ModelMetricClaimedVerifiedStatus
1ViT-B/16Average mAP60.17Unverified
2ResNet-50Average mAP56.19Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy51.2Unverified
2ZLaP*Accuracy51Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy29.1Unverified
2ZLaP*Accuracy29Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy75.9Unverified
2ZLaP*Accuracy75.5Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy87.9Unverified
2ZLaPAccuracy87.8Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPTop 1 Accuracy72.1Unverified
2ZLaP*Top 1 Accuracy72.1Unverified
#ModelMetricClaimedVerifiedStatus
1HiTeAAccuracy21.7Unverified
2HiTeAAccuracy0.46Unverified
#ModelMetricClaimedVerifiedStatus
1HiTeAAccuracy37.4Unverified
2HiTeAAccuracy0.56Unverified
#ModelMetricClaimedVerifiedStatus
1SPOTaverage top-1 classification accuracy71.9Unverified
2ZSL_TF-VAEGANaverage top-1 classification accuracy70.8Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaPAccuracy90Unverified
2ZLaP*Accuracy89Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy71.8Unverified
2ZLaPAccuracy71.2Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy71.4Unverified
2ZLaPAccuracy71Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy76.3Unverified
2ZLaPAccuracy76.3Unverified
#ModelMetricClaimedVerifiedStatus
1CLIP(ViT-B/16)Average mAP85.77Unverified
2CLIP(ResNet-50)Average mAP84.3Unverified
#ModelMetricClaimedVerifiedStatus
1ZSL-KGTop-160.54Unverified
#ModelMetricClaimedVerifiedStatus
1zsl_ADAAverage Per-Class Accuracy70.9Unverified
#ModelMetricClaimedVerifiedStatus
1ZLaP*Accuracy63.2Unverified
#ModelMetricClaimedVerifiedStatus
1MSDAPearson correlation coefficient (PCC)0.52Unverified
#ModelMetricClaimedVerifiedStatus
1SeViLAAccuracy72.3Unverified
#ModelMetricClaimedVerifiedStatus
1M^2-EncoderAccuracy80.7Unverified
#ModelMetricClaimedVerifiedStatus
1FrozenBiLMAccuracy51.5Unverified
#ModelMetricClaimedVerifiedStatus
1CZSLA-acc36Unverified
#ModelMetricClaimedVerifiedStatus
1ZS3Netk=10 mIOU26.3Unverified
#ModelMetricClaimedVerifiedStatus
1ZSL-KGAccuracy88.98Unverified
#ModelMetricClaimedVerifiedStatus
1VideoChat2Accuracy40.6Unverified