conftrace_

dataset overlap

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

train-test overlap (2) winograd schema challenge (1) pronoun disambiguation (1) open-domain question answering (1) commonsense reasoning (1) dataset evaluation (1) pretraining corpus (1) data leakage (1) language model (1) model evaluation (1)

Papers

Memorization vs. Generalization : Quantifying Data Leakage in NLP Performance Evaluation EACL 2021 Question and Answer Test-Train Overlap in Open-Domain Question Answering Datasets EACL 2021 An Analysis of Dataset Overlap on Winograd-Style Tasks COLING 2020