SOTAVerified

Memorization

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 676–700 of 1088 papers

TitleStatusHype
Memorization of Named Entities in Fine-tuned BERT ModelsCode0
Codex Hacks HackerRank: Memorization Issues and a Framework for Code Synthesis Evaluation—0
CrossSplit: Mitigating Label Noise Memorization through Data Splitting—0
Neural Representations Reveal Distinct Modes of Class Fitting in Residual Convolutional NetworksCode0
The Myth of Culturally Agnostic AI Models—0
Boundary Adversarial Examples Against Adversarial Overfitting—0
Data Origin Inference in Machine LearningCode0
Validating Large Language Models with ReLMCode1
DS-1000: A Natural and Reliable Benchmark for Data Science Code GenerationCode2
Why Deep Learning GeneralizesCode0
Prompting Language Models for Linguistic Structure—0
Selective Memory Recursive Least Squares: Recast Forgetting into Memory in RBF Neural Network Based Real-Time Learning—0
Multi-center anatomical segmentation with heterogeneous labels via landmark-based modelsCode1
Robust Training of Graph Neural Networks via Noise GovernanceCode1
Unintended Memorization and Timing Attacks in Named Entity Recognition Models—0
PLATO-K: Internal and External Knowledge Enhanced Dialogue Generation—0
Preventing Verbatim Memorization in Language Models Gives a False Sense of Privacy—0
Deep Learning is Provably Robust to Symmetric Label Noise—0
MemoNet: Memorizing All Cross Features' Representations Efficiently via Multi-Hash Codebook Network for CTR PredictionCode1
Empowering Long-tail Item Recommendation through Cross Decoupling Network (CDN)—0
The Curious Case of Benign Memorization—0
Finding Memo: Extractive Memorization in Constrained Sequence Generation TasksCode0
Reducing Training Sample Memorization in GANs by Training with Memorization RejectionCode0
Measures of Information Reflect Memorization Patterns—0
CNTN: Cyclic Noise-tolerant Network for Gait Recognition—0
Show:102550
← PrevPage 28 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