activity
20182020
most citedTransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agents

282 citations · 293 across the 2 of their papers we have counts for

collaborators

6 papers

cs.CL202011 cited

TLDR: Token Loss Dynamic Reweighting for Reducing Repetitive Utterance Generation

Shaojie Jiang, Thomas Wolf, Christof Monz +1

Natural Language Generation (NLG) models are prone to generating repetitive utterances. In this work, we study the repetition problem for encoder-decoder models, using both recurre…

cond-mat.str-el2019

Ultrafast magnetic dynamics in insulating YBaCuO revealed by time resolved two-magnon Raman Scattering

Jhih-An Yang, Nicholas Pellatz, Thomas Wolf +2

Measurement and control of magnetic order and correlations in real time is a rapidly developing scientific area relevant for magnetic memory and spintronics [1,15]. In these experi…

cs.CL2019

DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter

Victor Sanh, Lysandre Debut, Julien Chaumond +1

As Transfer Learning from large-scale pre-trained models becomes more prevalent in Natural Language Processing (NLP), operating these large models in on-the-edge and/or under const…

cond-mat.supr-con2019

Competing Electronic Phases near the Onset of Superconductivity in Hole-doped SrFe2As2

L. Wang, M. He, D. D. Scherer +6

An intriguingly complex phase diagram of Na-doped SrFe2As2 is uncovered using high-resolution thermal-expansion, magnetization and heat-capacity measurements. The detailed temperat…

cs.CL2019282 cited

TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agents

Thomas Wolf, Victor Sanh, Julien Chaumond +1

We introduce a new approach to generative data-driven dialogue systems (e.g. chatbots) called TransferTransfo which is a combination of a Transfer learning based training scheme an…

cs.CL2018

A Hierarchical Multi-task Approach for Learning Embeddings from Semantic Tasks

Victor Sanh, Thomas Wolf, Sebastian Ruder

Much effort has been devoted to evaluate whether multi-task learning can be leveraged to learn rich representations that can be used in various Natural Language Processing (NLP) do…