activity
20192022
most citedLook Who's Talking: Inferring Speaker Attributes from Personal Longitudinal Dialog

3 citations · 5 across the 5 of their papers we have counts for

collaborators

8 papers

cs.CL2022

Nearest Neighbor Language Models for Stylistic Controllable Generation

Severino Trotta, Lucie Flek, Charles Welch

Recent language modeling performance has been greatly improved by the use of external memory. This memory encodes the context so that similar contexts can be recalled during decodi…

cs.CL20222 cited

Mitigating Toxic Degeneration with Empathetic Data: Exploring the Relationship Between Toxicity and Empathy

Allison Lahnala, Charles Welch, Béla Neuendorf +1

Large pre-trained neural language models have supported the effectiveness of many NLP tasks, yet are still prone to generating toxic language hindering the safety of their use. Usi…

cs.CL2021

Modeling Proficiency with Implicit User Representations

Kim Breitwieser, Allison Lahnala, Charles Welch +2

We introduce the problem of proficiency modeling: Given a user's posts on a social media platform, the task is to identify the subset of posts or topics for which the user has some…

cs.CL2020

Exploring the Value of Personalized Word Embeddings

Charles Welch, Jonathan K. Kummerfeld, Verónica Pérez-Rosas +1

In this paper, we introduce personalized word embeddings, and examine their value for language modeling. We compare the performance of our proposed prediction model when using pers…

cs.CL2020

Compositional Demographic Word Embeddings

Charles Welch, Jonathan K. Kummerfeld, Verónica Pérez-Rosas +1

Word embeddings are usually derived from corpora containing text from many individuals, thus leading to general purpose representations rather than individually personalized repres…

cs.CL2020

Improving Low Compute Language Modeling with In-Domain Embedding Initialisation

Charles Welch, Rada Mihalcea, Jonathan K. Kummerfeld

Many NLP applications, such as biomedical data and technical support, have 10-100 million tokens of in-domain data and limited computational resources for learning from it. How sho…