diff --git a/t/real/rei.t b/t/real/rei.t
index 4ab9937..5e99df4 100644
--- a/t/real/rei.t
+++ b/t/real/rei.t
@@ -3,10 +3,7 @@
 use Test::More;
 use Data::Dumper;
 use JSON::XS;
-
-use Benchmark qw/:hireswallclock/;
-
-my $t = Benchmark->new;
+use Log::Log4perl;
 
 use utf8;
 use lib 'lib', '../lib';
@@ -14,6 +11,14 @@
 use File::Basename 'dirname';
 use File::Spec::Functions 'catdir';
 
+# Initialize log4perl object
+Log::Log4perl->init({
+  'log4perl.rootLogger' => 'ERROR, STDERR',
+  'log4perl.appender.STDERR' => 'Log::Log4perl::Appender::ScreenColoredLevels',
+  'log4perl.appender.STDERR.layout' => 'PatternLayout',
+  'log4perl.appender.STDERR.layout.ConversionPattern' => '[%r] %F %L %c - %m%n'
+});
+
 use_ok('KorAP::XML::Krill');
 
 my $path = catdir(dirname(__FILE__), '../corpus/REI/BNG/00128');
@@ -147,5 +152,176 @@
 like($first_token, qr/_0\$<i>0<i>8/, 'data');
 
 
+# REI/RBR/00610
+$path = catdir(dirname(__FILE__), '../corpus/REI/RBR/00610');
+
+ok($doc = KorAP::XML::Krill->new( path => $path . '/' ), 'Load Korap::Document');
+ok($doc->parse, 'Parse document');
+
+# Get tokenization
+$tokens = KorAP::XML::Tokenizer->new(
+  path => $doc->path,
+  doc => $doc,
+  foundry => $token_base_foundry,
+  layer => $token_base_layer,
+  name => 'tokens'
+);
+
+ok($tokens, 'Token Object is fine');
+ok($tokens->parse, 'Token parsing is fine');
+
+# Add annotations
+$tokens->add('Base', 'Sentences');
+$tokens->add('Base', 'Paragraphs');
+
+$tokens->add('CoreNLP', 'Constituency');
+$tokens->add('CoreNLP', 'Morpho');
+$tokens->add('CoreNLP', 'Sentences');
+
+$tokens->add('Malt', 'Dependency');
+
+$tokens->add('OpenNLP', 'Morpho');
+$tokens->add('OpenNLP', 'Sentences');
+
+$tokens->add('DeReKo', 'Structure');
+
+$tokens->add('TreeTagger', 'Morpho');
+$tokens->add('TreeTagger', 'Sentences');
+
+$output = decode_json( $tokens->to_json );
+
+my $first = $output->{data}->{stream}->[0];
+is('-:base/paragraphs$<i>31', $first->[0]);
+is('-:base/sentences$<i>168', $first->[1]);
+is('-:corenlp/sentences$<i>175', $first->[2]);
+is('-:opennlp/sentences$<i>169', $first->[3]);
+is('-:tokens$<i>2641', $first->[4]);
+is('-:tt/sentences$<i>196', $first->[5]);
+is('<:malt/d:PP$<b>32<i>4', $first->[6]);
+is('<>:corenlp/c:ROOT$<b>64<i>0<i>48<i>7<b>0', $first->[7]);
+is('<>:corenlp/s:s$<b>64<i>0<i>48<i>7<b>0', $first->[8]);
+is('<>:opennlp/s:s$<b>64<i>0<i>48<i>7<b>0', $first->[9]);
+is('<>:tt/s:s$<b>64<i>0<i>48<i>7<b>0', $first->[10]);
+is('<>:corenlp/c:S$<b>64<i>0<i>48<i>7<b>1', $first->[11]);
+is('<>:dereko/s:front$<b>64<i>0<i>91<i>11<b>1', $first->[12]);
+is('<>:base/s:s$<b>64<i>0<i>91<i>11<b>2', $first->[13]);
+is('<>:dereko/s:titlePage$<b>64<i>0<i>91<i>11<b>2<s>1', $first->[14]);
+is('<>:dereko/s:docTitle$<b>64<i>0<i>91<i>11<b>3', $first->[15]);
+is('<>:dereko/s:titlePart$<b>64<i>0<i>91<i>11<b>4<s>2', $first->[16]);
+is('<>:dereko/s:s$<b>64<i>0<i>91<i>11<b>5', $first->[17]);
+is('<>:base/s:t$<b>64<i>0<i>17859<i>2640<b>0', $first->[18]);
+is('>:malt/d:ROOT$<b>33<i>0<i>48<i>0<i>7', $first->[19]);
+is('<:malt/d:PP$<b>32<i>1', $first->[20]);
+is('<:malt/d:ROOT$<b>34<i>0<i>48<i>7<i>0', $first->[21]);
+is('@:dereko/s:id:rbr.00610-0-titlepage$<b>17<s>1<i>11', $first->[22]);
+is('@:dereko/s:type:unspecified$<b>17<s>2<i>11', $first->[23]);
+is('_0$<i>0<i>4', $first->[24]);
+is('corenlp/p:NN', $first->[25]);
+is('i:rede', $first->[26]);
+is('opennlp/p:NN', $first->[27]);
+is('s:Rede', $first->[28]);
+is('tt/l:Rede$<b>129<b>253', $first->[29]);
+is('tt/p:NN$<b>129<b>253', $first->[30]);
+
+my $last = $output->{data}->{stream}->[-1];
+
+is('<>:dereko/s:text$<b>65<i>17859<i>17859<i>2640<b>0', $last->[0]);
+is('<>:dereko/s:back$<b>65<i>17859<i>17859<i>2640<b>1', $last->[1]);
+is('>:malt/d:APP$<b>32<i>2639', $last->[2]);
+is('_2640$<i>17851<i>17859', $last->[3]);
+is('corenlp/p:NE', $last->[4]);
+is("i:schr\x{f6}der", $last->[5]);
+is('opennlp/p:NE', $last->[6]);
+is("s:Schr\x{f6}der", $last->[7]);
+is("tt/l:Schr\x{f6}der\$<b>129<b>255", $last->[8]);
+is('tt/p:NE$<b>129<b>255', $last->[9]);
+
+
+# REI/BNG/00071
+$path = catdir(dirname(__FILE__), '../corpus/REI/BNG/00071');
+
+ok($doc = KorAP::XML::Krill->new( path => $path . '/' ), 'Load Korap::Document');
+ok($doc->parse, 'Parse document');
+
+# Get tokenization
+$tokens = KorAP::XML::Tokenizer->new(
+  path => $doc->path,
+  doc => $doc,
+  foundry => $token_base_foundry,
+  layer => $token_base_layer,
+  name => 'tokens'
+);
+
+ok($tokens, 'Token Object is fine');
+ok($tokens->parse, 'Token parsing is fine');
+
+# Add annotations
+$tokens->add('Base', 'Sentences');
+$tokens->add('Base', 'Paragraphs');
+
+$tokens->add('CoreNLP', 'Constituency');
+$tokens->add('CoreNLP', 'Morpho');
+$tokens->add('CoreNLP', 'Sentences');
+
+$tokens->add('Malt', 'Dependency');
+
+$tokens->add('OpenNLP', 'Morpho');
+$tokens->add('OpenNLP', 'Sentences');
+
+$tokens->add('DeReKo', 'Structure');
+
+$tokens->add('TreeTagger', 'Morpho');
+$tokens->add('TreeTagger', 'Sentences');
+
+$output = decode_json( $tokens->to_json );
+
+$first = $output->{data}->{stream}->[0];
+
+is('-:base/paragraphs$<i>41', $first->[0]);
+is('-:base/sentences$<i>73', $first->[1]);
+is('-:corenlp/sentences$<i>69', $first->[2]);
+is('-:opennlp/sentences$<i>65', $first->[3]);
+is('-:tokens$<i>1009', $first->[4]);
+is('-:tt/sentences$<i>95', $first->[5]);
+is('<:malt/d:APP$<b>32<i>1', $first->[6]);
+is('<>:corenlp/c:MPN$<b>64<i>0<i>16<i>2<b>3', $first->[7]);
+is('<>:base/s:s$<b>64<i>0<i>48<i>5<b>2', $first->[8]);
+is('<>:dereko/s:titlePart$<b>64<i>0<i>48<i>5<b>4<s>2', $first->[9]);
+is('<>:dereko/s:s$<b>64<i>0<i>48<i>5<b>5', $first->[10]);
+is('<>:corenlp/c:ROOT$<b>64<i>0<i>51<i>6<b>0', $first->[11]);
+is('<>:corenlp/s:s$<b>64<i>0<i>51<i>6<b>0', $first->[12]);
+is('<>:tt/s:s$<b>64<i>0<i>51<i>6<b>0', $first->[13]);
+is('<>:corenlp/c:NUR$<b>64<i>0<i>51<i>6<b>1', $first->[14]);
+is('<>:corenlp/c:NP$<b>64<i>0<i>50<i>6<b>2', $first->[15]);
+is('<>:dereko/s:front$<b>64<i>0<i>61<i>8<b>1', $first->[16]);
+is('<>:dereko/s:titlePage$<b>64<i>0<i>61<i>8<b>2<s>1', $first->[17]);
+is('<>:dereko/s:docTitle$<b>64<i>0<i>61<i>8<b>3', $first->[18]);
+is('<>:opennlp/s:s$<b>64<i>0<i>173<i>24<b>0', $first->[19]);
+is('<>:base/s:t$<b>64<i>0<i>7008<i>1008<b>0', $first->[20]);
+is('<>:dereko/s:text$<b>64<i>0<i>7008<i>1008<b>0', $first->[21]);
+is('>:malt/d:GMOD$<b>32<i>3', $first->[22]);
+is('<:malt/d:ROOT$<b>34<i>0<i>51<i>6<i>3', $first->[23]);
+is('@:dereko/s:id:bng.00071-0-titlepage$<b>17<s>1<i>8', $first->[24]);
+is('@:dereko/s:type:unspecified$<b>17<s>2<i>5', $first->[25]);
+is('_0$<i>0<i>9', $first->[26]);
+is('corenlp/p:NE', $first->[27]);
+is('i:christine', $first->[28]);
+is('opennlp/p:NE', $first->[29]);
+is('s:Christine', $first->[30]);
+is('tt/l:Christine$<b>129<b>255', $first->[31]);
+is('tt/p:NE$<b>129<b>255', $first->[32]);
+
+$last = $output->{data}->{stream}->[-1];
+is('<>:dereko/s:back$<b>65<i>7008<i>7008<i>1008<b>1', $last->[0]);
+is('<>:dereko/s:div$<b>65<i>7008<i>7008<i>1008<b>2<s>1', $last->[1]);
+is('@:dereko/s:n:1$<b>17<s>1', $last->[2]);
+is('@:dereko/s:type:footnotes$<b>17<s>1', $last->[3]);
+is('@:dereko/s:complete:y$<b>17<s>1', $last->[4]);
+is('_1008$<i>6990<i>7006', $last->[5]);
+is('corenlp/p:NN', $last->[6]);
+is('i:befreiungsschlag', $last->[7]);
+is('opennlp/p:NN', $last->[8]);
+is('s:Befreiungsschlag', $last->[9]);
+
 done_testing;
 __END__