Search Results for author: Dan Moldovan

Found 19 papers, 4 papers with code

A Tool for Extracting Conversational Implicatures

no code implementations LREC 2012 Marta Tatu, Dan Moldovan

Explicitly conveyed knowledge represents only a portion of the information communicated by a text snippet.

Common Sense Reasoning Implicatures +1

Generating Questions for Reading Comprehension using Coherence Relations

no code implementations WS 2018 Takshak Desai, Parag Dakle, Dan Moldovan

In this paper, we have proposed a technique for generating complex reading comprehension questions from a discourse that are more useful than factual ones derived from assertions.

Reading Comprehension

Rule-based vs. Neural Net Approaches to Semantic Textual Similarity

no code implementations COLING 2018 Linrui Zhang, Dan Moldovan

This paper presents a neural net approach to determine Semantic Textual Similarity (STS) using attention-based bidirectional Long Short-Term Memory Networks (Bi-LSTM).

Feature Engineering Semantic Textual Similarity +2

Tangent: Automatic differentiation using source-code transformation for dynamically typed array programming

no code implementations NeurIPS 2018 Bart van Merriënboer, Dan Moldovan, Alexander B. Wiltschko

The need to efficiently calculate first- and higher-order derivatives of increasingly complex models expressed in Python has stressed or exceeded the capabilities of available tools.

AutoGraph: Imperative-style Coding with Graph-based Performance

no code implementations16 Oct 2018 Dan Moldovan, James M Decker, Fei Wang, Andrew A Johnson, Brian K. Lee, Zachary Nado, D. Sculley, Tiark Rompf, Alexander B. Wiltschko

In machine learning, imperative style libraries like Autograd and PyTorch are easy to write, but suffer from high interpretive overhead and are not easily deployable in production or mobile settings.

BIG-bench Machine Learning

Joint Learning of Syntactic Features Helps Discourse Segmentation

1 code implementation LREC 2020 Takshak Desai, Parag Pravin Dakle, Dan Moldovan

This paper describes an accurate framework for carrying out multi-lingual discourse segmentation with BERT (Devlin et al., 2019).

Discourse Segmentation token-classification

A Study on Entity Resolution for Email Conversations

no code implementations LREC 2020 Parag Pravin Dakle, Takshak Desai, Dan Moldovan

This paper investigates the problem of entity resolution for email conversations and presents a seed annotated corpus of email threads labeled with entity coreference chains.

Entity Resolution

Affect inTweets: A Transfer Learning Approach

no code implementations LREC 2020 Linrui Zhang, Hsin-Lun Huang, Yang Yu, Dan Moldovan

As opposed to the traditional machine learning models which require considerable effort in designing task specific features, our model can be well adapted to the proposed tasks with a very limited amount of fine-tuning, which significantly reduces the manual effort in feature engineering.

Feature Engineering Transfer Learning

Causally motivated Shortcut Removal Using Auxiliary Labels

1 code implementation13 May 2021 Maggie Makar, Ben Packer, Dan Moldovan, Davis Blalock, Yoni Halpern, Alexander D'Amour

Shortcut learning, in which models make use of easy-to-represent but unstable associations, is a major failure mode for robust machine learning.

Causal Inference Disentanglement +1

Cannot find the paper you are looking for? You can Submit a new open access paper.