Search Results for author: Greg Durrett

Found 53 papers, 28 papers with code

Dissecting Generation Modes for Abstractive Summarization Models via Ablation and Attribution

1 code implementation3 Jun 2021 Jiacheng Xu, Greg Durrett

Despite the prominence of neural abstractive summarization models, we know little about how they actually form summaries and how to understand where their decisions come from.

Abstractive Text Summarization Language Modelling +1

Can NLI Models Verify QA Systems' Predictions?

no code implementations18 Apr 2021 Jifan Chen, Eunsol Choi, Greg Durrett

To build robust question answering systems, we need the ability to verify whether answers to questions are truly correct, not just "good enough" in the context of imperfect QA datasets.

Natural Language Inference Question Answering

Flexible Operations for Natural Language Deduction

1 code implementation18 Apr 2021 Kaj Bostrom, Xinyu Zhao, Swarat Chaudhuri, Greg Durrett

Natural language is an excellent candidate -- it is both extremely expressive and easy for humans to understand.

Evaluating Explanations for Reading Comprehension with Realistic Counterfactuals

1 code implementation9 Apr 2021 Xi Ye, Rohan Nair, Greg Durrett

Token-level attributions have been extensively studied to explain model predictions for a wide range of classification tasks in NLP (e. g., sentiment analysis), but such explanation techniques are less explored for machine reading comprehension (RC) tasks.

Machine Reading Comprehension Sentiment Analysis

Annotating and Modeling Fine-grained Factuality in Summarization

1 code implementation NAACL 2021 Tanya Goyal, Greg Durrett

Recent pre-trained abstractive summarization systems have started to achieve credible performance, but a major barrier to their use in practice is their propensity to output summaries that are not faithful to the input and that contain factual errors.

Abstractive Text Summarization

Did they answer? Subjective acts and intents in conversational discourse

1 code implementation NAACL 2021 Elisa Ferracane, Greg Durrett, Junyi Jessy Li, Katrin Erk

Discourse signals are often implicit, leaving it up to the interpreter to draw the required inferences.

Model Agnostic Answer Reranking System for Adversarial Question Answering

no code implementations EACL 2021 Sagnik Majumder, Chinmoy Samant, Greg Durrett

While numerous methods have been proposed as defenses against adversarial examples in question answering (QA), these techniques are often model specific, require retraining of the model, and give only marginal improvements in performance over vanilla models.

Question Answering

Modeling Fine-Grained Entity Types with Box Embeddings

1 code implementation2 Jan 2021 Yasumasa Onoe, Michael Boratko, Andrew McCallum, Greg Durrett

Neural entity typing models typically represent fine-grained entity types as vectors in a high-dimensional space, but such spaces are not well-suited to modeling these types' complex interdependencies.

Entity Typing

Conditional Generation of Temporally-ordered Event Sequences

no code implementations31 Dec 2020 Shih-ting Lin, Nathanael Chambers, Greg Durrett

We propose a single model that addresses both temporal ordering, sorting given events into the order they occurred, and event infilling, predicting new events which fit into an existing temporally-ordered sequence.

Denoising Story Completion

Effective Distant Supervision for Temporal Relation Extraction

1 code implementation24 Oct 2020 Xinyu Zhao, Shih-ting Lin, Greg Durrett

A principal barrier to training temporal relation extraction models in new domains is the lack of varied, high quality examples and the challenge of collecting more.

Relation Extraction

Compressive Summarization with Plausibility and Salience Modeling

1 code implementation EMNLP 2020 Shrey Desai, Jiacheng Xu, Greg Durrett

Compressive summarization systems typically rely on a crafted set of syntactic rules to determine what spans of possible summary sentences can be deleted, then learn a model of what to actually delete by optimizing for content selection (ROUGE).

Understanding Neural Abstractive Summarization Models via Uncertainty

1 code implementation EMNLP 2020 Jiacheng Xu, Shrey Desai, Greg Durrett

An advantage of seq2seq abstractive summarization models is that they generate text in a free-form manner, but this flexibility makes it difficult to interpret model behavior.

Abstractive Text Summarization Text Generation

Evaluating Factuality in Generation with Dependency-level Entailment

1 code implementation Findings of the Association for Computational Linguistics 2020 Tanya Goyal, Greg Durrett

Experiments show that our dependency arc entailment model trained on this data can identify factual inconsistencies in paraphrasing and summarization better than sentence-level methods or those based on question generation, while additionally localizing the erroneous parts of the generation.

Natural Language Inference Question Generation +1

Inquisitive Question Generation for High Level Text Comprehension

1 code implementation EMNLP 2020 Wei-Jen Ko, Te-Yuan Chen, Yiyan Huang, Greg Durrett, Junyi Jessy Li

Inquisitive probing questions come naturally to humans in a variety of settings, but is a challenging task for automatic systems.

Question Generation Reading Comprehension

Optimal Neural Program Synthesis from Multimodal Specifications

no code implementations4 Oct 2020 Xi Ye, Qiaochu Chen, Isil Dillig, Greg Durrett

Multimodal program synthesis, which leverages different types of user input to synthesize a desired program, is an attractive way to scale program synthesis to challenging settings; however, it requires integrating noisy signals from the user (like natural language) with hard constraints on the program's behavior.

Program Synthesis

Tradeoffs in Sentence Selection Techniques for Open-Domain Question Answering

no code implementations18 Sep 2020 Shih-ting Lin, Greg Durrett

Current methods in open-domain question answering (QA) usually employ a pipeline of first retrieving relevant documents, then applying strong reading comprehension (RC) models to that retrieved text.

Open-Domain Question Answering Reading Comprehension

Narrative Interpolation for Generating and Understanding Stories

no code implementations17 Aug 2020 Su Wang, Greg Durrett, Katrin Erk

We propose a method for controlled narrative/story generation where we are able to guide the model to produce coherent narratives with user-specified target endings by interpolation: for example, we are told that Jim went hiking and at the end Jim needed to be rescued, and we want the model to incrementally generate steps along the way.

Neural Syntactic Preordering for Controlled Paraphrase Generation

2 code implementations ACL 2020 Tanya Goyal, Greg Durrett

Paraphrasing natural language sentences is a multifaceted process: it might involve replacing individual words or short phrases, local rearrangement of content, or high-level restructuring like topicalization or passivization.

Machine Translation Paraphrase Generation

Benchmarking Multimodal Regex Synthesis with Complex Structures

no code implementations ACL 2020 Xi Ye, Qiaochu Chen, Isil Dillig, Greg Durrett

Existing datasets for regular expression (regex) generation from natural language are limited in complexity; compared to regex tasks that users post on StackOverflow, the regexes in these datasets are simple, and the language used to describe them is not diverse.

Robust Question Answering Through Sub-part Alignment

no code implementations NAACL 2021 Jifan Chen, Greg Durrett

Current textual question answering models achieve strong performance on in-domain test sets, but often do so by fitting surface-level patterns in the data, so they fail to generalize to out-of-distribution settings.

Question Answering

Interpretable Entity Representations through Large-Scale Typing

no code implementations Findings of the Association for Computational Linguistics 2020 Yasumasa Onoe, Greg Durrett

On entity probing tasks involving recognizing entity identity, our embeddings used in parameter-free downstream models achieve competitive performance with ELMo- and BERT-based embeddings in trained models.

Entity Embeddings Entity Typing

LambdaNet: Probabilistic Type Inference using Graph Neural Networks

1 code implementation ICLR 2020 Jiayi Wei, Maruth Goyal, Greg Durrett, Isil Dillig

Given this program abstraction, we then use a graph neural network to propagate information between related type variables and eventually make type predictions.

Byte Pair Encoding is Suboptimal for Language Model Pretraining

no code implementations Findings of the Association for Computational Linguistics 2020 Kaj Bostrom, Greg Durrett

We analyze differences between BPE and unigram LM tokenization, finding that the latter method recovers subword units that align more closely with morphology and avoids problems stemming from BPE's greedy construction procedure.

Language Modelling Tokenization

Calibration of Pre-trained Transformers

2 code implementations EMNLP 2020 Shrey Desai, Greg Durrett

Pre-trained Transformers are now ubiquitous in natural language processing, but despite their high end-task performance, little is known empirically about whether they are calibrated.

Natural Language Inference

Multi-hop Question Answering via Reasoning Chains

3 code implementations7 Oct 2019 Jifan Chen, Shih-ting Lin, Greg Durrett

Our analysis shows the properties of chains that are crucial for high performance: in particular, modeling extraction sequentially is important, as is dealing with each candidate sentence in a context-aware way.

Coreference Resolution Multi-hop Question Answering +2

Query-Focused Scenario Construction

no code implementations IJCNLP 2019 Su Wang, Greg Durrett, Katrin Erk

The news coverage of events often contains not one but multiple incompatible accounts of what happened.

Fine-Grained Entity Typing for Domain Independent Entity Linking

1 code implementation12 Sep 2019 Yasumasa Onoe, Greg Durrett

For this problem, a domain is characterized not just by genre of text but even by factors as specific as the particular distribution of entities, as neural models tend to overfit by memorizing properties of frequent entities in a dataset.

Entity Linking Entity Typing

Effective Use of Transformer Networks for Entity Tracking

1 code implementation IJCNLP 2019 Aditya Gupta, Greg Durrett

Tracking entities in procedural language requires understanding the transformations arising from actions on entities as well as those entities' interactions.

Natural Language Understanding

Sketch-Driven Regular Expression Generation from Natural Language and Examples

1 code implementation16 Aug 2019 Xi Ye, Qiaochu Chen, Xinyu Wang, Isil Dillig, Greg Durrett

Our system achieves state-of-the-art performance on the prior datasets and solves 57% of the real-world dataset, which existing neural systems completely fail on.

Embedding time expressions for deep temporal ordering models

3 code implementations ACL 2019 Tanya Goyal, Greg Durrett

Data-driven models have demonstrated state-of-the-art performance in inferring the temporal ordering of events in text.

Learning to Denoise Distantly-Labeled Data for Entity Typing

1 code implementation NAACL 2019 Yasumasa Onoe, Greg Durrett

In this work, we propose a two-stage procedure for handling this type of data: denoise it with a learned model, then train our final model on clean and denoised distant data with standard supervised training.

Denoising Entity Typing

Understanding Dataset Design Choices for Multi-hop Reasoning

no code implementations NAACL 2019 Jifan Chen, Greg Durrett

First, we explore sentence-factored models for these tasks; by design, these models cannot do multi-hop reasoning, but they are still able to solve a large number of examples in both datasets.

Multi-hop Question Answering Question Answering +1

Tracking Discrete and Continuous Entity State for Process Understanding

no code implementations WS 2019 Aditya Gupta, Greg Durrett

The global discrete state structure is explicitly modeled with a neural CRF over the changing hidden representation of the entity.

Domain Agnostic Real-Valued Specificity Prediction

1 code implementation13 Nov 2018 Wei-Jen Ko, Greg Durrett, Junyi Jessy Li

Sentence specificity quantifies the level of detail in a sentence, characterizing the organization of information in discourse.

Dialogue Generation Unsupervised Domain Adaptation

Picking Apart Story Salads

no code implementations EMNLP 2018 Su Wang, Eric Holgate, Greg Durrett, Katrin Erk

During natural disasters and conflicts, information about what happened is often confusing, messy, and distributed across many sources.

Effective Use of Context in Noisy Entity Linking

1 code implementation EMNLP 2018 David Mueller, Greg Durrett

To disambiguate between closely related concepts, entity linking systems need to effectively distill cues from their context, which may be quite noisy.

Entity Linking

Spherical Latent Spaces for Stable Variational Autoencoders

1 code implementation EMNLP 2018 Jiacheng Xu, Greg Durrett

A hallmark of variational autoencoders (VAEs) for text processing is their combination of powerful encoder-decoder models, such as LSTMs, with simple latent distributions, typically multivariate Gaussians.

Language Modelling

Modeling Semantic Plausibility by Injecting World Knowledge

1 code implementation NAACL 2018 Su Wang, Greg Durrett, Katrin Erk

Distributional data tells us that a man can swallow candy, but not that a man can swallow a paintball, since this is never attested.

Capturing Semantic Similarity for Entity Linking with Convolutional Neural Networks

1 code implementation NAACL 2016 Matthew Francis-Landau, Greg Durrett, Dan Klein

A key challenge in entity linking is making effective use of contextual information to disambiguate mentions that might refer to different entities in different contexts.

Entity Linking Semantic correspondence +2

Learning-Based Single-Document Summarization with Compression and Anaphoricity Constraints

no code implementations ACL 2016 Greg Durrett, Taylor Berg-Kirkpatrick, Dan Klein

We present a discriminative model for single-document summarization that integrally combines compression and anaphoricity constraints.

Document Summarization

Neural CRF Parsing

no code implementations IJCNLP 2015 Greg Durrett, Dan Klein

This paper describes a parsing model that combines the exact dynamic programming of CRF parsing with the rich nonlinear featurization of neural net approaches.

A Joint Model for Entity Analysis: Coreference, Typing, and Linking

no code implementations TACL 2014 Greg Durrett, Dan Klein

We present a joint model of three core tasks in the entity analysis stack: coreference resolution (within-document clustering), named entity recognition (coarse semantic typing), and entity linking (matching to Wikipedia entities).

Coreference Resolution Entity Linking +1

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