David Alvarez-Melis

Topic Modeling in Twitter: Aggregating Tweets by Conversations

Word Embeddings as Metric Recovery in Semantic Spaces

Tree-structured decoding with doubly-recurrent neural networks

A causal framework for explaining the predictions of black-box sequence-to-sequence models.

Word, graph and manifold embedding from Markov processes.

Distributional Adversarial Networks

Structured Optimal Transport.

Gromov-Wasserstein Alignment of Word Embedding Spaces

On the Robustness of Interpretability Methods.

Game-Theoretic Interpretability for Temporal Modeling.