Humans often have to read multiple documents to address their information needs.However, most existing reading comprehension (RC) tasks only focus on questions for which the contexts provide all the information required to answer them, thus not evaluating a system's performance at identifying a potential lack of sufficient information and locating sources for that information.To fill this gap, we present a dataset, IIRC, with more than 13K questions over paragraphs from English Wikipedia that provide only partial information to answer them, with the missing information occurring in one or more linked documents.The questions were written by crowd workers who did not have access to any of the linked documents, leading to questions that have little lexical overlap with the contexts where the answers appear.This process also gave many questions without answers, and those that require discrete reasoning, increasing the difficulty of the task.We follow recent modeling work on various reading comprehension datasets to construct a baseline model for this dataset, finding that it achieves 31.1% F1 on this task, while estimated human performance is 88.4%.The dataset, code for the baseline system, and a leaderboard can be found at https://allennlp.org/iirc.

When the semantics of a sentence are not representable in a semantic parser's output schema, parsing will inevitably fail.Detection of these instances is commonly treated as an out-of-domain classification problem.However, there is also a more subtle scenario in which the test data is drawn from the same domain.In addition to formalizing this problem of domain-adjacency, we present a comparison of various baselines that could be used to solve it.We also propose a new simple sentence representation that emphasizes words which are unexpected.This approach improves the performance of a downstream semantic parser run on in-domain and domainadjacent instances.

We present a discriminative model for detecting disfluencies in spoken language transcripts.Structurally, our model is a semi-Markov conditional random field with features targeting characteristics unique to speech repairs.This gives a significant performance improvement over standard chain-structured CRFs that have been employed in past work.We then incorporate prosodic features over silences and relative word duration into our semi-CRF model, resulting in further performance gains; moreover, these features are not easily replaced by discrete prosodic indicators such as ToBI breaks.Our final system, the semi-CRF with prosodic information, achieves an F-score of 85.4, which is 1.3 F 1 better than the best prior reported F-score on this dataset.

The problem of defining a smooth surface through an array of points in space is well known. Several methods of solution have been proposed. Generally, these restrict the set of points to be one-to-one defined over a planar rectangular grid ( X , Y -plane). Then a set of functions Z = F ( X , Y ) is determined, each of which represents a surface segment of the composite smooth surface. In this paper, these ideas are generalized to include a much broader class of permissible point array distributions: namely (1) the point arrangement (ordering) is topologically equivalent to a planar rectangular grid, (2) the resulting solution is a smooth composite of parametric surface segments, i.e. each surface piece is represented by a vector (point)-valued function. The solution here presented is readily applicable to a variety of problems, such as closed surface body definitions and pressure envelope surface definitions. The technique has been used successfully in these areas and others, such as numerical control milling, Newtonian impact and boundary layer.