合并分支 'caorunzhe' 到 'master'

Caorunzhe

查看合并请求 !885

Chapter8/Figures/figure-cky-algorithm.tex

@@ -28,7 +28,7 @@
 
 
 
 
 \begin{pgfonlayer}{background}
 \begin{pgfonlayer}{background}
-\node [rectangle,inner sep=0.2em,rounded corners=1pt,fill=blue!10!white] [fit = (c1) (c21) (c3) (c6) (c7) (c8) (c11)] (gl1) {};
+\node [rectangle,draw=ublue,thick,inner sep=0.2em,fill=white,drop shadow] [fit = (c1) (c21) (c3) (c6) (c7) (c8) (c11)] (gl1) {};
 \end{pgfonlayer}
 \end{pgfonlayer}
 
 
 \end{scope}
 \end{scope}

Chapter8/Figures/figure-classification-of-models-based-on-syntax.tex

@@ -4,23 +4,23 @@
 \begin{tikzpicture}
 \begin{tikzpicture}
 
 
 \begin{scope}
 \begin{scope}
-\tikzstyle{cnode} = [minimum width=7.0em,minimum height=2.5em,rounded corners=0.2em];
-\tikzstyle{xnode} = [minimum width=4.5em,minimum height=2.5em,rounded corners=0.2em];
+\tikzstyle{cnode} = [minimum width=7.0em,minimum height=2.5em,rounded corners=0.2em,draw,thick];
+\tikzstyle{xnode} = [minimum width=4.5em,minimum height=2.5em,rounded corners=0.2em,draw,thick];
 
 
 \node[cnode,anchor=south,minimum width=10.0em,fill=green!25,align=center] (cat0) at (0,0) {\footnotesize{（广义上）}\\\footnotesize{基于句法的模型}};
 \node[cnode,anchor=south,minimum width=10.0em,fill=green!25,align=center] (cat0) at (0,0) {\footnotesize{（广义上）}\\\footnotesize{基于句法的模型}};
-\node[cnode,anchor=north,fill=red!25,align=left] (cat1) at ([xshift=-6.5em,yshift=-2em]cat0.south) {\footnotesize{基于形式文法}\\\footnotesize{的模型}};
-\node[cnode,anchor=north,fill=blue!25,align=left] (cat2) at ([xshift=6.5em,yshift=-2em]cat0.south) {\footnotesize{基于语言学}\\\footnotesize{句法的模型}};
+\node[cnode,anchor=north,fill=red!25,align=left] (cat1) at ([xshift=-7.5em,yshift=-2em]cat0.south) {\footnotesize{基于形式文法}\\\footnotesize{的模型}};
+\node[cnode,anchor=north,fill=blue!25,align=left] (cat2) at ([xshift=7.5em,yshift=-2em]cat0.south) {\footnotesize{基于语言学}\\\footnotesize{句法的模型}};
 
 
 
 
-\node[xnode,anchor=north,fill=red!25,align=left] (itg) at ([xshift=-2.5em,yshift=-2.0em]cat1.south) {\footnotesize{反向转录}\\\footnotesize{文法}};
-\node[xnode,anchor=north,fill=red!25,align=left] (hiero) at ([xshift=2.5em,yshift=-2.0em]cat1.south) {\footnotesize{层次短语}\\\footnotesize{模型}};
-\node[xnode,anchor=north,fill=blue!25,align=left] (s2t) at ([xshift=-5.0em,yshift=-2.0em]cat2.south) {\footnotesize{串到树}\\\footnotesize{模型}};
+\node[xnode,anchor=north,fill=red!25,align=left] (itg) at ([xshift=-3.5em,yshift=-2.0em]cat1.south) {\footnotesize{反向转录}\\\footnotesize{文法}};
+\node[xnode,anchor=north,fill=red!25,align=left] (hiero) at ([xshift=3.5em,yshift=-2.0em]cat1.south) {\footnotesize{层次短语}\\\footnotesize{模型}};
+\node[xnode,anchor=north,fill=blue!25,align=left] (s2t) at ([xshift=-5.5em,yshift=-2.0em]cat2.south) {\footnotesize{串到树}\\\footnotesize{模型}};
 \node[xnode,anchor=north,fill=blue!25,align=left] (t2s) at ([xshift=0.0em,yshift=-2.0em]cat2.south) {\footnotesize{树到串}\\\footnotesize{模型}};
 \node[xnode,anchor=north,fill=blue!25,align=left] (t2s) at ([xshift=0.0em,yshift=-2.0em]cat2.south) {\footnotesize{树到串}\\\footnotesize{模型}};
-\node[xnode,anchor=north,fill=blue!25,align=left] (t2t) at ([xshift=5.0em,yshift=-2.0em]cat2.south) {\footnotesize{树到树}\\\footnotesize{模型}};
+\node[xnode,anchor=north,fill=blue!25,align=left] (t2t) at ([xshift=5.5em,yshift=-2.0em]cat2.south) {\footnotesize{树到树}\\\footnotesize{模型}};
 
 
 
 
-\draw [-,thick] ([yshift=0.1em,xshift=1em]cat0.south) -- ([xshift=-1.5em,yshift=-0.1em]cat2.north);
-\draw [-,thick] ([yshift=0.1em,xshift=-1em]cat0.south) -- ([xshift=1.5em,yshift=-0.1em]cat1.north);
+\draw [-,thick] ([yshift=-0.1em,xshift=1em]cat0.south) -- ([xshift=-1.5em,yshift=0.1em]cat2.north);
+\draw [-,thick] ([yshift=-0.1em,xshift=-1em]cat0.south) -- ([xshift=1.5em,yshift=0.1em]cat1.north);
 \draw [-,thick] ([yshift=0.1em]itg.north) -- ([xshift=-0.5em,yshift=-0.1em]cat1.south);
 \draw [-,thick] ([yshift=0.1em]itg.north) -- ([xshift=-0.5em,yshift=-0.1em]cat1.south);
 \draw [-,thick] ([yshift=0.1em]hiero.north) -- ([xshift=0.5em,yshift=-0.1em]cat1.south);
 \draw [-,thick] ([yshift=0.1em]hiero.north) -- ([xshift=0.5em,yshift=-0.1em]cat1.south);
 \draw [-,thick] ([yshift=0.1em]s2t.north) -- ([xshift=-0.8em,yshift=-0.1em]cat2.south);
 \draw [-,thick] ([yshift=0.1em]s2t.north) -- ([xshift=-0.8em,yshift=-0.1em]cat2.south);

Chapter8/Figures/figure-derivation-of-hierarchical-phrase-and-tree-structure-model.tex

@@ -39,7 +39,7 @@
 }
 }
 \end{scope}
 \end{scope}
 
 
-\begin{scope}[xshift = 1.8in, yshift = 0.1in]
+\begin{scope}[xshift = 2.0in, yshift = 0.1in]
 \node (rules) {\textbf{层次短语翻译规则:}};
 \node (rules) {\textbf{层次短语翻译规则:}};
 \draw[-] (rules.south west)--([xshift=1.8in]rules.south west);
 \draw[-] (rules.south west)--([xshift=1.8in]rules.south west);
 
 

Chapter8/Figures/figure-different-representations-of-syntax-tree.tex

@@ -6,16 +6,16 @@
 \Tree[.S [.NN ] [.VP [.AD ] [.VP [.VV ] [.AS ] ] ] ]
 \Tree[.S [.NN ] [.VP [.AD ] [.VP [.VV ] [.AS ] ] ] ]
 \end{scope}
 \end{scope}
 	
 	
-\begin{scope}[xshift=1.25in,yshift=0.25in]
+\begin{scope}[xshift=1.5in,yshift=0.25in]
 \node [anchor=north west,align=left] (string1) at (0,0) {[S \\\hspace{1em}NN\\\hspace{1em}VP[\\\hspace{2.5em}AD \\\hspace{2.5em}VP[\\\hspace{4em}VV \\\hspace{4em}AS]]]};
 \node [anchor=north west,align=left] (string1) at (0,0) {[S \\\hspace{1em}NN\\\hspace{1em}VP[\\\hspace{2.5em}AD \\\hspace{2.5em}VP[\\\hspace{4em}VV \\\hspace{4em}AS]]]};
 \end{scope}
 \end{scope}
 	
 	
-\begin{scope}[xshift=2.78in,yshift=-0.80in]
+\begin{scope}[xshift=3.3in,yshift=-0.80in]
 \node [anchor=west,align=left] (string2) at (0,0) {(S NN VP(AD \\ VP(VV AS)))};
 \node [anchor=west,align=left] (string2) at (0,0) {(S NN VP(AD \\ VP(VV AS)))};
 \end{scope}
 \end{scope}
 	
 	
 \node [anchor=north west] (cap1) at (-1.5em,-1in) {{(a) 树状表示}};
 \node [anchor=north west] (cap1) at (-1.5em,-1in) {{(a) 树状表示}};
-\node [anchor=west] (cap2) at ([xshift=0.5in]cap1.east) {{(b) 序列表示（缩进）}};
-\node [anchor=west] (cap3) at ([xshift=0.3in]cap2.east) {{(c) 序列表示}};
+\node [anchor=west] (cap2) at ([xshift=0.7in]cap1.east) {{(b) 序列表示（缩进）}};
+\node [anchor=west] (cap3) at ([xshift=0.7in]cap2.east) {{(c) 序列表示}};
 }
 }
 \end{tikzpicture}
 \end{tikzpicture}
\ No newline at end of file

Chapter8/Figures/figure-processing-of-hierarchical-phrase-system.tex

@@ -4,9 +4,9 @@
 \begin{tikzpicture}
 \begin{tikzpicture}
 \begin{scope}
 \begin{scope}
 
 
-\tikzstyle{datanode} = [minimum width=7em,minimum height=1.7em,fill=blue!20,rounded corners=0.3em];
-\tikzstyle{modelnode} = [minimum width=7em,minimum height=1.7em,fill=red!20,rounded corners=0.3em];
-\tikzstyle{decodingnode} = [minimum width=7em,minimum height=1.7em,fill=green!20,rounded corners=0.3em];
+\tikzstyle{datanode} = [minimum width=7em,minimum height=1.7em,fill=blue!20,rounded corners=0.3em,draw,thick];
+\tikzstyle{modelnode} = [minimum width=7em,minimum height=1.7em,fill=red!20,rounded corners=0.3em,draw,thick];
+\tikzstyle{decodingnode} = [minimum width=7em,minimum height=1.7em,fill=green!20,rounded corners=0.3em,draw,thick];
 
 
 \node [datanode,anchor=north west,minimum height=1.7em,minimum width=8em] (bitext) at (0,0) {{ \small{训练用双语数据}}};
 \node [datanode,anchor=north west,minimum height=1.7em,minimum width=8em] (bitext) at (0,0) {{ \small{训练用双语数据}}};
 \node [modelnode, anchor=north west,minimum height=1.7em,minimum width=8em] (gi) at ([xshift=2em,yshift=-0.2em]bitext.south east) {{ \small{文法（规则）抽取}}};
 \node [modelnode, anchor=north west,minimum height=1.7em,minimum width=8em] (gi) at ([xshift=2em,yshift=-0.2em]bitext.south east) {{ \small{文法（规则）抽取}}};
@@ -21,18 +21,18 @@
 \node [modelnode,anchor=south west,minimum height=1.7em,minimum width=8em] (lm) at ([xshift=2em,yshift=0.2em]tuning.north east) {{ \small{$n$-gram语言建模}}};
 \node [modelnode,anchor=south west,minimum height=1.7em,minimum width=8em] (lm) at ([xshift=2em,yshift=0.2em]tuning.north east) {{ \small{$n$-gram语言建模}}};
 \node [datanode,anchor=south west,minimum height=1.7em,minimum width=8em] (lmmodel) at ([xshift=2em,yshift=0.2em]decoding.north east) {{ \small{语言模型}}};
 \node [datanode,anchor=south west,minimum height=1.7em,minimum width=8em] (lmmodel) at ([xshift=2em,yshift=0.2em]decoding.north east) {{ \small{语言模型}}};
 
 
-\draw [->,very thick] ([yshift=-0.3em,xshift=0.1em]bitext.east) -- ([yshift=0.1em]gi.north west);
-\draw [->,very thick] ([yshift=-0.1em]gi.south west) -- ([yshift=0.3em,xshift=0.1em]birules.east);
-\draw [->,very thick] ([yshift=-0.3em,xshift=0.1em]birules.east) -- ([yshift=0.1em]training.north west);
-\draw [->,very thick] ([yshift=-0.1em]training.south west) -- ([yshift=0.3em,xshift=0.1em]model.east);
-\draw [->,very thick] ([yshift=-0.3em,xshift=0.1em]model.east) -- ([yshift=0.1em]tuning.north west);
-\draw [->,very thick] ([yshift=-0.1em]tuning.south) -- ([yshift=0.1em]decoding.north);
-\draw [->,very thick] ([yshift=0.3em,xshift=0.1em]tuningdata.east) -- ([yshift=-0.1em]tuning.south west);
+\draw [->,thick] ([yshift=-0.3em,xshift=0.1em]bitext.east) -- ([yshift=0.1em]gi.north west);
+\draw [->,thick] ([yshift=-0.1em]gi.south west) -- ([yshift=0.3em,xshift=0.1em]birules.east);
+\draw [->,thick] ([yshift=-0.3em,xshift=0.1em]birules.east) -- ([yshift=0.1em]training.north west);
+\draw [->,thick] ([yshift=-0.1em]training.south west) -- ([yshift=0.3em,xshift=0.1em]model.east);
+\draw [->,thick] ([yshift=-0.3em,xshift=0.1em]model.east) -- ([yshift=0.1em]tuning.north west);
+\draw [->,thick] ([yshift=-0.1em]tuning.south) -- ([yshift=0.1em]decoding.north);
+\draw [->,thick] ([yshift=0.3em,xshift=0.1em]tuningdata.east) -- ([yshift=-0.1em]tuning.south west);
 
 
-\draw [->,very thick] ([yshift=-0.1em]monotext.south) -- ([yshift=0.1em]lm.north);
-\draw [->,very thick] ([yshift=-0.1em]lm.south) -- ([yshift=0.1em]lmmodel.north);
-\draw [->,very thick] ([yshift=0.3em,xshift=-0.1em]lmmodel.west) -- ([yshift=-0.1em]tuning.south east);
-\draw [->,very thick] ([yshift=-0.3em,xshift=-0.1em]lmmodel.west) -- ([yshift=0.1em]decoding.north east);
+\draw [->,thick] ([yshift=-0.1em]monotext.south) -- ([yshift=0.1em]lm.north);
+\draw [->,thick] ([yshift=-0.1em]lm.south) -- ([yshift=0.1em]lmmodel.north);
+\draw [->,thick] ([yshift=0.3em,xshift=-0.1em]lmmodel.west) -- ([yshift=-0.1em]tuning.south east);
+\draw [->,thick] ([yshift=-0.3em,xshift=-0.1em]lmmodel.west) -- ([yshift=0.1em]decoding.north east);
 \end{scope}
 \end{scope}
 \end{tikzpicture}
 \end{tikzpicture}
 \end{center}
 \end{center}

Chapter8/Figures/figure-translation-rule-describe-two-sentence-generation.tex

@@ -32,8 +32,8 @@
 
 
 \begin{pgfonlayer}{background}
 \begin{pgfonlayer}{background}
 {
 {
-\node [rectangle,draw=red,inner sep=1pt,thick,fill=white,drop shadow] [fit = (rule1s1) (rule1s4)] (rule1s) {};
-\node [rectangle,draw=red,inner sep=2pt,thick,fill=white,drop shadow] [fit = (rule1t1) (rule1t4)] (rule1t) {};
+\node [rectangle,draw=red,inner sep=1pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule1s1) (rule1s4)] (rule1s) {};
+\node [rectangle,draw=red,inner sep=2pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule1t1) (rule1t4)] (rule1t) {};
 \draw [<->,dotted,thick,red] ([xshift=0.1em]rule1s.east) -- ([xshift=-0.1em]rule1t.west);
 \draw [<->,dotted,thick,red] ([xshift=0.1em]rule1s.east) -- ([xshift=-0.1em]rule1t.west);
 }
 }
 \end{pgfonlayer}
 \end{pgfonlayer}
@@ -70,8 +70,8 @@
 
 
 \begin{pgfonlayer}{background}
 \begin{pgfonlayer}{background}
 {
 {
-\node [rectangle,draw=blue,inner sep=4pt,thick,fill=white,drop shadow] [fit = (rule2s1) (rule2s3)] (rule2s) {};
-\node [rectangle,draw=blue,inner sep=4pt,thick,fill=white,drop shadow] [fit = (rule2t1) (rule2t3)] (rule2t) {};
+\node [rectangle,draw=blue,inner sep=4pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule2s1) (rule2s3)] (rule2s) {};
+\node [rectangle,draw=blue,inner sep=4pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule2t1) (rule2t3)] (rule2t) {};
 \draw [<->,dotted,thick,blue] ([xshift=0.1em]rule2s.east) -- ([xshift=-0.1em]rule2t.west);
 \draw [<->,dotted,thick,blue] ([xshift=0.1em]rule2s.east) -- ([xshift=-0.1em]rule2t.west);
 }
 }
 \end{pgfonlayer}
 \end{pgfonlayer}
@@ -92,8 +92,8 @@
 
 
 \begin{pgfonlayer}{background}
 \begin{pgfonlayer}{background}
 {
 {
-\node [rectangle,draw=ugreen,inner sep=1pt,thick,fill=white,drop shadow] [fit = (rule3s1) (rule3s2)] (rule3s) {};
-\node [rectangle,draw=ugreen,inner sep=2pt,thick,fill=white,drop shadow] [fit = (rule3t1) (rule3t2)] (rule3t) {};
+\node [rectangle,draw=ugreen,inner sep=1pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule3s1) (rule3s2)] (rule3s) {};
+\node [rectangle,draw=ugreen,inner sep=2pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule3t1) (rule3t2)] (rule3t) {};
 \draw [<->,dotted,thick,ugreen] ([xshift=0.1em]rule3s.east) -- ([xshift=-0.1em]rule3t.west);
 \draw [<->,dotted,thick,ugreen] ([xshift=0.1em]rule3s.east) -- ([xshift=-0.1em]rule3t.west);
 }
 }
 \end{pgfonlayer}
 \end{pgfonlayer}
@@ -122,8 +122,8 @@
 
 
 \begin{pgfonlayer}{background}
 \begin{pgfonlayer}{background}
 {
 {
-\node [rectangle,draw=orange,inner sep=4pt,thick,fill=white,drop shadow] [fit = (rule4s1) (rule4s3)] (rule4s) {};
-\node [rectangle,draw=orange,inner sep=4pt,thick,fill=white,drop shadow] [fit = (rule4t1) (rule4t3)] (rule4t) {};
+\node [rectangle,draw=orange,inner sep=4pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule4s1) (rule4s3)] (rule4s) {};
+\node [rectangle,draw=orange,inner sep=4pt,thick,fill=white,drop shadow,rounded corners=0.3em] [fit = (rule4t1) (rule4t3)] (rule4t) {};
 \draw [<->,dotted,thick,orange] ([xshift=0.1em]rule4s.east) -- ([xshift=-0.1em]rule4t.west);
 \draw [<->,dotted,thick,orange] ([xshift=0.1em]rule4s.east) -- ([xshift=-0.1em]rule4t.west);
 }
 }
 \end{pgfonlayer}
 \end{pgfonlayer}

Chapter8/Figures/figure-tree-cutting-defined-by-edge-nodes.tex

@@ -65,7 +65,7 @@
 
 
 \end{scope}
 \end{scope}
 
 
-\begin{scope} [yshift = -1.87in, xshift = 1.9in]
+\begin{scope} [yshift = -1.87in, xshift = 2.2in]
 {
 {
 \begin{scope}[level distance=20pt]
 \begin{scope}[level distance=20pt]
 \node[scale=0.8, inner sep=0.1cm,align=center,draw] (cfrag1) at
 \node[scale=0.8, inner sep=0.1cm,align=center,draw] (cfrag1) at

bibliography.bib

@@ -9331,6 +9331,20 @@ author    = {Zhuang Liu and
   publisher = {Conference on Empirical Methods in Natural Language Processing},
   publisher = {Conference on Empirical Methods in Natural Language Processing},
   year      = {2016}
   year      = {2016}
 }
 }
+@inproceedings{tu2017neural,
+  title={Neural machine translation with reconstruction},
+  author={Tu, Zhaopeng and Liu, Yang and Shang, Lifeng and Liu, Xiaohua and Li, Hang},
+  publisher={AAAI Conference on Artificial Intelligence},
+  volume={31},
+  number={1},
+  year={2017}
+}
+@inproceedings{li2020train,
+  title={Train large, then compress: Rethinking model size for efficient training and inference of transformers},
+  author={Li, Zhuohan and Wallace, Eric and Shen, Sheng and Lin, Kevin and Keutzer, Kurt and Klein, Dan and Gonzalez, Joseph E},
+  publisher={arXiv preprint arXiv:2002.11794},
+  year={2020}
+}
 %%%%% chapter 15------------------------------------------------------
 %%%%% chapter 15------------------------------------------------------
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%