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.
Natural language is an excellent candidate -- it is both extremely expressive and easy for humans to understand.
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.
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.
Discourse signals are often implicit, leaving it up to the interpreter to draw the required inferences.
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.
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.
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.
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.
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).
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.
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.
Inquisitive probing questions come naturally to humans in a variety of settings, but is a challenging task for automatic systems.
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.
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.
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.
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.
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.
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.
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.
Given this program abstraction, we then use a graph neural network to propagate information between related type variables and eventually make type predictions.
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.
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.
Ranked #3 on Question Answering on WikiHop
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.
Tracking entities in procedural language requires understanding the transformations arising from actions on entities as well as those entities' interactions.
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.
Data-driven models have demonstrated state-of-the-art performance in inferring the temporal ordering of events in text.
Discourse structure is integral to understanding a text and is helpful in many NLP tasks.
Sequence-to-sequence models for open-domain dialogue generation tend to favor generic, uninformative responses.
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.
Ranked #1 on Entity Typing on Ontonotes v5 (English)
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.
The global discrete state structure is explicitly modeled with a neural CRF over the changing hidden representation of the entity.
In this work, we present a neural model for single-document summarization based on joint extraction and syntactic compression.
Ranked #1 on Extractive Text Summarization on CNN / Daily Mail
Sentence specificity quantifies the level of detail in a sentence, characterizing the organization of information in discourse.
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.
Distributional data tells us that a man can swallow candy, but not that a man can swallow a paintball, since this is never attested.
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.
We present a discriminative model for single-document summarization that integrally combines compression and anaphoricity constraints.
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).
Ranked #19 on Named Entity Recognition on Ontonotes v5 (English)