105 research outputs found
Dependency-based n-gram models for general purpose sentence realisation
No full textWe present dependency-based n-gram models for general-purpose, widecoverage, probabilistic sentence realisation.
Our method linearises unordered dependencies in input representations directly rather than via the application
of grammar rules, as in traditional chartbased generators. The method is simple, efficient, and achieves competitive accuracy and complete coverage on standard English (Penn-II, 0.7440 BLEU, 0.05 sec/sent) and Chinese (CTB6, 0.7123 BLEU, 0.14 sec/sent) test data
Accurate and robust LFG-based generation for Chinese
No full textWe describe three PCFG-based models for Chinese sentence realisation from Lexical-Functional Grammar (LFG) f-structures. Both the lexicalised model and the history-based model improve on the accuracy of a simple
wide-coverage PCFG model by adding lexical and contextual information to weaken inappropriate independence assumptions implicit in the PCFG models. In addition, we provide techniques for lexical smoothing and rule smoothing to increase the generation coverage. Trained on 15,663 automatically LFG fstructure annotated sentences of the Penn Chinese treebank and tested on 500 sentences randomly
selected from the treebank test set, the lexicalised model achieves a BLEU score of 0.7265 at 100% coverage, while the historybased model achieves a BLEU score of 0.7245
also at 100% coverage
CCG-YOLOv7: A Wood Defect Detection Model for Small Targets Using Improved YOLOv7
No full textThe Chinese furniture market has a high demand for wood floors. Manual defect detection in wood floors is inefficient and lacks stability in accuracy. It is necessary to conduct research on automatic defect detection in wood floors. To improve the accuracy of detecting small defects in wood floors, this paper proposed a new network based on YOLOv7. The new network is called the cascade center of gravity YOLOv7 (CCG-YOLOv7). This paper designed cascade efficient layer aggregation networks (C-ELAN), streamlined the CBS, replaced the ELAN with the C-ELAN, introduced the rapid supervised attention module to connect the backbone and head layers, and simplified the head layer of the YOLOv7 network. These methods improved the detection accuracy and speed for detecting small defects on wood floor surfaces. The improved network can effectively detect small defects on the wooden board surfaces, including knots, scratches, and mildew. Compared to the original YOLOv7, CCG-YOLOv7 improves precision, recall, and mean average precision by 2.1%, 1.6%, and 1.2%, respectively
Formation mechanism of coarse columnar γ grains in as-cast hyperperitectic carbon steels
No full textThe formation mechanism of as-cast coarse columnar γ grain (CCG) structure in hyperperitectic carbon steels is investigated by means of rapid unidirectional solidification method. This method realizes the cooling conditions similar to those in the vicinity of a practical continuously cast slab surface. The microstructural observation of the quenched samples indicates that the CCG structure develops from the mold side along the direction of the temperature gradient. In the solidifying samples, fine columnar γ grains (FCGs) always exist ahead of the CCG region. Instead of continuous growth into the large grains, the FCGs always shrink and vanish due to the growth of CCGs initially formed near the mold side. Therefore, the grain size at a fixed point of the ingot discontinuously changes from the FCG to the CCG. The validity of this process was supported by numerical analyses. This finding is in marked contrast to the assumption made in conventional grain growth analysis on the CCG structure
A critical condition for the formation of a coarse columnar gamma grain structure in a peritectic solidified carbon steel
No full textThe formation of a coarse columnar austenite grain (CCG) structure is a serious problem in continuous casting processes of peritectic solidified carbon steels. In this study, a guiding principle for the avoidance of CCG formation is developed. The critical condition for CCG formation recently put forward based on phase-field simulations, which is given by a balance between the cooling condition and the growth rate of the CCG, is first re-examined and modified by considering the effect of a liquid phase during the CCG formation. The validity of this critical condition is then investigated by three different casting experiments combined with heat conduction analyses. From a comparison between the cooling conditions and the resulting microstructural changes, the validity of the critical condition is successfully demonstrated. (C) 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved
Treebank-based acquisition of LFG parsing resources for French
No full textMotivated by the expense in time and other resources to produce hand-crafted grammars, there has been increased interest in automatically obtained wide-coverage grammars from treebanks for natural language processing. In particular, recent years have seen the growth in interest in automatically obtained deep resources that can represent information absent from simple CFG-type structured treebanks
and which are considered to produce more language-neutral linguistic representations, such as dependency syntactic trees. As is often the case in early pioneering work on natural language processing, English has provided the focus of first efforts towards acquiring deep-grammar resources, followed by successful treatments of, for example, German, Japanese, Chinese and Spanish. However, no comparable large-scale automatically acquired deep-grammar resources have been obtained for French to date. The goal of this paper is to present the application of treebank-based language acquisition to the case of French. We show that with modest changes to the established parsing architectures, encouraging results can be obtained for French, with a best dependency structure f-score of 86.73%
Experimental Verification of a Critical Condition for the Formation of As-Cast Coarse Columnar Austenite Grain Structure in a Hyperperitectic Carbon Steel
No full textExperimental verification of a critical condition for the formation of coarse columnar gamma grain (CCG) structure in as-cast hyperperitectic carbon steels, which was put forward based on theories of grain growth and phase-field simulations in early studies, is carried out by means of a Bridgman-type directional solidification experiment. The occurrence of the discontinuous and continuous grain growth processes and the resulting formation of CCG and equiaxed gamma grain structures, respectively, are demonstrated. Importantly, these changes of the as-cast microstructures and the grain growth modes are in excellent agreement with the previously proposed critical condition of the CCG formation. (C) The Minerals, Metals & Materials Society and ASM International 201
For illustration, one central slice of the T1-weighted template data (T1w), maps and <i>φ</i> maps are selectively shown with all four anatomical regions, i.e., the head of caudate nucleus (HCN), thalamus (TH), corpus callosum genu (CCG), and white matter (WM).
No full text<p>Regions of HCN, TH, and CCG were manually selected while the WM region was automatically segmented from the T1w.</p
Formation conditions of coarse columnar austenite grain structure in peritectic carbon steels by the discontinuous grain growth mechanism
No full textThe discontinuous grain growth leading to a coarse columnar austenite grain (CCG) structure in as-cast peritectic carbon steels is analyzed by means of two- and three-dimensional phase-field simulations. On the basis of theory of grain growth, the conditions for discontinuous grain growth to occur is elucidated in terms of cooling condition and material properties. The theoretical analysis well explains the results of the phase-field simulations and it should be useful in predicting the formation of the CCG structure in continuously-cast slabs
Divergence in Dialogue
No full textCopyright: 2014 Healey et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.This work was supported by the Economic and Social Research Council (ESRC; http://www.esrc.ac.uk/) through the DynDial project (Dynamics of Conversational Dialogue, RES-062-23-0962) and the Engineering and Physical Sciences Research Council (EPSRC; http://www.epsrc.ac.uk/) through the RISER
project (Robust Incremental Semantic Resources for Dialogue, EP/J010383/1). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript
- …
