activity
20132026
most citedAligning where to see and what to tell: image caption with region-based attention and scene factorization

107 citations · 287 across the 30 of their papers we have counts for

collaborators
Showing 2021Show all

6 papers · 1 filter

cs.CL2021

Learning to Generalize Compositionally by Transferring Across Semantic Parsing Tasks

Wang Zhu, Peter Shaw, Tal Linzen +1

Neural network models often generalize poorly to mismatched domains or distributions. In NLP, this issue arises in particular when models are expected to generalize compositionally…

cs.LG2021★ 12 cited

HyperPINN: Learning parameterized differential equations with physics-informed hypernetworks

Filipe de Avila Belbute-Peres, Yi-fan Chen, Fei Sha

Many types of physics-informed neural network models have been proposed in recent years as approaches for learning solutions to differential equations. When a particular task requi…

cs.CL2021★ 1 cited

Systematic Generalization on gSCAN: What is Nearly Solved and What is Next?

Linlu Qiu, Hexiang Hu, Bowen Zhang +2

We analyze the grounded SCAN (gSCAN) benchmark, which was recently proposed to study systematic generalization for grounded language understanding. First, we study which aspects of…

cs.CL2021

ReadTwice: Reading Very Large Documents with Memories

Yury Zemlyanskiy, Joshua Ainslie, Michiel de Jong +3

Knowledge-intensive tasks such as question answering often require assimilating information from different sections of large inputs such as books or article collections. We propose…

cs.LG2021★ 5 cited

Embedding Adaptation is Still Needed for Few-Shot Learning

Sébastien M. R. Arnold, Fei Sha

Constructing new and more challenging tasksets is a fruitful methodology to analyse and understand few-shot classification methods. Unfortunately, existing approaches to building t…

cs.CL2021

DOCENT: Learning Self-Supervised Entity Representations from Large Document Collections

Yury Zemlyanskiy, Sudeep Gandhe, Ruining He +5

This paper explores learning rich self-supervised entity representations from large amounts of the associated text. Once pre-trained, these models become applicable to multiple ent…