• Publications
  • Influence
AdapterFusion: Non-Destructive Task Composition for Transfer Learning
TLDR
We propose AdapterFusion, a new two stage learning algorithm that leverages knowledge from multiple tasks in a non-destructive manner. Expand
AdapterHub: A Framework for Adapting Transformers
TLDR
We propose AdapterHub, a framework that allows dynamic “stiching-in” of pre-trained adapters for different tasks and languages, particularly in low-resource scenarios. Expand
A Survey on Semantic Parsing
TLDR
A significant amount of information in today's world is stored in structured and semi-structured knowledge bases. Expand
MDETR - Modulated Detection for End-to-End Multi-Modal Understanding
TLDR
We propose an end-to-end modulated detector that detects objects in an image conditioned on a raw text query, like a caption or a question. Expand
Specializing Distributional Vectors of All Words for Lexical Entailment
TLDR
We present the first post-processing method that specializes vectors of all vocabulary words – including those unseen in the resources – for the asymmetric relation of lexical entailment (LE) (i.e., hyponymy-hypernymy relation). Expand
What do Deep Networks Like to Read?
TLDR
We propose Susceptibility Identification through Fine-Tuning (SIFT), a novel abstractive method that uncovers a model's preferences without imposing any prior. Expand
Training Structured Prediction Energy Networks with Indirect Supervision
TLDR
This paper introduces rank-based training of structured prediction energy networks using gradient descent and minimizes the ranking violation of the sampled structures with respect to a scalar scoring function defined with domain knowledge. Expand