Search Results for author: Nicholas Botzer

Found 4 papers, 3 papers with code

TK-KNN: A Balanced Distance-Based Pseudo Labeling Approach for Semi-Supervised Intent Classification

1 code implementation17 Oct 2023 Nicholas Botzer, David Vasquez, Tim Weninger, Issam Laradji

In the present work, we describe Top-K K-Nearest Neighbor (TK-KNN), which uses a more robust pseudo-labeling approach based on distance in the embedding space while maintaining a balanced set of pseudo-labeled examples across classes through a ranking-based approach.

intent-classification Intent Classification

Posthoc Verification and the Fallibility of the Ground Truth

1 code implementation NAACL (DADC) 2022 Yifan Ding, Nicholas Botzer, Tim Weninger

Metrics used in these evaluations are tied to the availability of well-defined ground truth labels, and these metrics typically do not allow for inexact matches.

Entity Linking

Reddit Entity Linking Dataset

no code implementations4 Jan 2021 Nicholas Botzer, Yifan Ding, Tim Weninger

We introduce and make publicly available an entity linking dataset from Reddit that contains 17, 316 linked entities, each annotated by three human annotators and then grouped into Gold, Silver, and Bronze to indicate inter-annotator agreement.

Entity Linking

HetSeq: Distributed GPU Training on Heterogeneous Infrastructure

1 code implementation25 Sep 2020 Yifan Ding, Nicholas Botzer, Tim Weninger

The present work describes HetSeq, a software package adapted from the popular PyTorch package that provides the capability to train large neural network models on heterogeneous infrastructure.

Image Classification Language Modelling +1

Cannot find the paper you are looking for? You can Submit a new open access paper.