SOTAVerified

Memorization

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 951–975 of 1088 papers

TitleStatusHype
Weak and Strong Gradient Directions: Explaining Memorization, Generalization, and Hardness of Examples at Scale—0
State-of-the-Art Augmented NLP Transformer models for direct and single-step retrosynthesisCode1
An Efficient Method of Training Small Models for Regression Problems with Knowledge Distillation—0
Do We Need Zero Training Loss After Achieving Zero Training Error?Code1
Improving Generalization by Controlling Label-Noise Information in Neural Network WeightsCode1
Learning Not to Learn in the Presence of Noisy Labels—0
Self-Attentive Associative MemoryCode1
A Corrective View of Neural Networks: Representation, Memorization and Learning—0
Encoding-based Memory Modules for Recurrent Neural Networks—0
EgoMap: Projective mapping and structured egocentric memory for Deep RL—0
Towards GAN Benchmarks Which Require Generalization—0
Online Memorization of Random Firing Sequences by a Recurrent Neural Network—0
Searching to Exploit Memorization Effect in Learning with Noisy Labels—0
Learning Human Postural Control with Hierarchical Acquisition Functions—0
The Labeling Distribution Matrix (LDM): A Tool for Estimating Machine Learning Algorithm Capacity—0
Variational Recurrent Models for Solving Partially Observable Control TasksCode0
Towards Robust Learning with Different Label Noise DistributionsCode0
Meta-Learning without MemorizationCode0
Semantic Mask for Transformer based End-to-End Speech RecognitionCode0
Neural Networks Learning and Memorization with (almost) no Over-Parameterization—0
Neural Network Memorization Dissection—0
Searching to Exploit Memorization Effect in Learning from Corrupted LabelsCode0
Generalization through Memorization: Nearest Neighbor Language ModelsCode1
On the Unintended Social Bias of Training Language Generation Models with Data from Local Media—0
Robust Training with Ensemble Consensus—0
Show:102550
← PrevPage 39 of 44Next →

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