no code implementations • BioNLP (ACL) 2022 • Zach Wood-Doughty, Isabel Cachola, Mark Dredze
We propose to use knowledge distillation, or training a student model that mimics the behavior of a trained teacher model, as a technique to generate faithful and plausible explanations.
1 code implementation • 29 Sep 2023 • Jingqian Wu, Rongtao Xu, Zach Wood-Doughty, Changwei Wang, Shibiao Xu, Edmund Lam
To do so, first, we construct an auxiliary task of Pixel Semantic Relational Distillation (PSRD), which distillates feature relations with category-agnostic semantic information learned by the SAM encoder into a local feature learning network, to improve local feature description using semantic discrimination.
1 code implementation • 2 Sep 2021 • Amir Feder, Katherine A. Keith, Emaad Manzoor, Reid Pryzant, Dhanya Sridhar, Zach Wood-Doughty, Jacob Eisenstein, Justin Grimmer, Roi Reichart, Margaret E. Roberts, Brandon M. Stewart, Victor Veitch, Diyi Yang
A fundamental goal of scientific research is to learn about causal relationships.
no code implementations • 15 Aug 2021 • Ilya Shpitser, Zach Wood-Doughty, Eric J. Tchetgen Tchetgen
Unobserved confounding is a fundamental obstacle to establishing valid causal conclusions from observational data.
1 code implementation • 16 Apr 2021 • Zach Wood-Doughty, Isabel Cachola, Mark Dredze
Machine learning models that offer excellent predictive performance often lack the interpretability necessary to support integrated human machine decision-making.
no code implementations • 10 Feb 2021 • Zach Wood-Doughty, Ilya Shpitser, Mark Dredze
High-dimensional and unstructured data such as natural language complicates the evaluation of causal inference methods; such evaluations rely on synthetic datasets with known causal effects.
no code implementations • 13 Oct 2020 • Aaron Mueller, Zach Wood-Doughty, Silvio Amir, Mark Dredze, Alicia L. Nobles
The #MeToo movement on Twitter has drawn attention to the pervasive nature of sexual harassment and violence.
1 code implementation • NAACL (SocialNLP) 2021 • Zach Wood-Doughty, Paiheng Xu, Xiao Liu, Mark Dredze
We present a method to identify self-reports of race and ethnicity from Twitter profile descriptions.
no code implementations • WS 2018 • Zach Wood-Doughty, Nicholas Andrews, Mark Dredze
While recurrent neural networks (RNNs) are widely used for text classification, they demonstrate poor performance and slow convergence when trained on long sequences.
1 code implementation • EMNLP 2018 • Zach Wood-Doughty, Ilya Shpitser, Mark Dredze
Causal understanding is essential for many kinds of decision-making, but causal inference from observational data has typically only been applied to structured, low-dimensional datasets.
1 code implementation • WS 2018 • Zach Wood-Doughty, Nicholas Andrews, Rebecca Marvin, Mark Dredze
Social media analysis frequently requires tools that can automatically infer demographics to contextualize trends.
1 code implementation • WS 2018 • Zach Wood-Doughty, Praateek Mahajan, Mark Dredze
Previous work (McCorriston et al., 2015) presented a method for determining if an account was an individual or organization based on account profile and a collection of tweets.
no code implementations • WS 2017 • Zach Wood-Doughty, Michael Smith, David Broniatowski, Mark Dredze
Demographically-tagged social media messages are a common source of data for computational social science.