activity
20182020
collaborators

7 papers

cs.CL2020

A Closer Look at Linguistic Knowledge in Masked Language Models: The Case of Relative Clauses in American English

Marius Mosbach, Stefania Degaetano-Ortlieb, Marie-Pauline Krielke +2

Transformer-based language models achieve high performance on various tasks, but we still lack understanding of the kind of linguistic knowledge they learn and rely on. We evaluate…

cs.CL2020

On the Interplay Between Fine-tuning and Sentence-level Probing for Linguistic Knowledge in Pre-trained Transformers

Marius Mosbach, Anna Khokhlova, Michael A. Hedderich +1

Fine-tuning pre-trained contextualized embedding models has become an integral part of the NLP pipeline. At the same time, probing has emerged as a way to investigate the linguisti…

cs.CV2020

Fusion Models for Improved Visual Captioning

Marimuthu Kalimuthu, Aditya Mogadala, Marius Mosbach +1

Visual captioning aims to generate textual descriptions given images or videos. Traditionally, image captioning models are trained on human annotated datasets such as Flickr30k and…

cs.CV2020

Sparse Graph to Sequence Learning for Vision Conditioned Long Textual Sequence Generation

Aditya Mogadala, Marius Mosbach, Dietrich Klakow

Generating longer textual sequences when conditioned on the visual information is an interesting problem to explore. The challenge here proliferate over the standard vision conditi…

cs.LG2020

On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines

Marius Mosbach, Maksym Andriushchenko, Dietrich Klakow

Fine-tuning pre-trained transformer-based language models such as BERT has become a common practice dominating leaderboards across various NLP benchmarks. Despite the strong empiri…

cs.CR2019

On the security relevance of weights in deep learning

Kathrin Grosse, Thomas A. Trost, Marius Mosbach +2

Recently, a weight-based attack on stochastic gradient descent inducing overfitting has been proposed. We show that the threat is broader: A task-independent permutation on the ini…