SOTAVerified

Memorization

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 851–900 of 1088 papers

TitleStatusHype
Exploring Memorization in Adversarial TrainingCode1
Learning to Rehearse in Long Sequence Memorization—0
Not All Knowledge Is Created Equal: Mutual Distillation of Confident Knowledge—0
Automatic Classification of Attributes in German Adjective-Noun PhrasesCode0
Evaluating and Explaining Natural Language Generation with GenXCode0
Understanding Unintended Memorization in Language Models Under Federated Learning—0
Do RNN States Encode Abstract Phonological Alternations?—0
Privacy Regularization: Joint Privacy-Utility Optimization in LanguageModels—0
On the geometry of generalization and memorization in deep neural networks—0
Annotation Inconsistency and Entity Bias in MultiWOZ—0
Using Early-Learning Regularization to Classify Real-World Noisy Data—0
Combining Recurrent, Convolutional, and Continuous-time Models with Linear State Space Layers—0
Learning to Generate Novel Scene Compositions from Single Images and VideosCode1
Bounding Information Leakage in Machine Learning—0
Self-paced Resistance Learning against Overfitting on Noisy LabelsCode0
Broccoli: Sprinkling Lightweight Vocabulary Learning into Everyday Information DietsCode0
Memorization vs. Generalization : Quantifying Data Leakage in NLP Performance Evaluation—0
Do RNN States Encode Abstract Phonological Processes?—0
Contrast to Divide: Self-Supervised Pre-Training for Learning with Noisy LabelsCode1
Jo-SRC: A Contrastive Approach for Combating Noisy Labels—0
Generating Novel Scene Compositions from Single Images and VideosCode1
Co-matching: Combating Noisy Labels by Augmentation Anchoring—0
On the Robustness of Monte Carlo Dropout Trained with Noisy Labels—0
Privacy Regularization: Joint Privacy-Utility Optimization in Language Models—0
On the privacy-utility trade-off in differentially private hierarchical text classificationCode0
Parsimonious Inference—0
Membership Inference Attacks are Easier on Difficult ProblemsCode0
Driving Style Representation in Convolutional Recurrent Neural Network Model of Driver IdentificationCode1
Robust Generalization and Safe Query-Specialization in Counterfactual Learning to RankCode0
Memorization vs. Generalization: Quantifying Data Leakage in NLP Performance EvaluationCode0
Learning to Combat Noisy Labels via Classification Margins—0
Meta-Regularization by Enforcing Mutual-ExclusivenessCode0
PConv: Simple yet Effective Convolutional Layer for Generative Adversarial Network—0
Hyperspectral Image Super-Resolution with Spectral Mixup and Heterogeneous DatasetsCode1
Investigating Memorization of Conspiracy Theories in Text GenerationCode0
Grounding Consistency: Distilling Spatial Common Sense for Precise Visual Relationship DetectionCode1
How Do Your Biomedical Named Entity Recognition Models Generalize to Novel Entities?Code0
Continual Memory: Can We Reason After Long-Term Memorization?—0
ME-MOMENTUM: EXTRACTING HARD CONFIDENT EXAMPLES FROM NOISILY LABELED DATA—0
Making Coherence Out of Nothing At All: Measuring Evolution of Gradient Alignment—0
Robust early-learning: Hindering the memorization of noisy labels—0
The Unreasonable Effectiveness of the Class-reversed Sampling in Tail Sample Memorization—0
Catching the Long Tail in Deep Neural Networks—0
A Large-scale Study on Training Sample Memorization in Generative Modeling—0
Distributed Associative Memory Network with Association Reinforcing Loss—0
Introducing Orthogonal Constraint in Structural ProbesCode0
Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization—0
Tight Bounds on the Smallest Eigenvalue of the Neural Tangent Kernel for Deep ReLU Networks—0
Perceptron Theory Can Predict the Accuracy of Neural Networks—0
When is Memorization of Irrelevant Training Data Necessary for High-Accuracy Learning?Code0
Show:102550
← PrevPage 18 of 22Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1PaLM-540B (few-shot, k=5)Accuracy95.4—Unverified
2Gopher-280B (few-shot, k=5)Accuracy80—Unverified
3PaLM-62B (few-shot, k=5)Accuracy77.7—Unverified