diff --git a/test/RunTests b/test/RunTests
index a66514db83b..cb1e28c8413 100755
--- a/test/RunTests
+++ b/test/RunTests
@@ -1052,7 +1052,7 @@ __DATA__
     train-sets/ref/search_er.stderr
 
 # Test 66: Train a depenency parser with search (dagger) on wsj_small.dparser.vw.gz for 2 passes
-{VW} -k -c -d train-sets/wsj_small.dparser.vw.gz --passes 2 --search_task dep_parser --search 3  --search_alpha 1e-4 --search_rollout oracle --holdout_off
+{VW} -k -c -d train-sets/wsj_small.dparser.vw.gz --passes 6 --search_task dep_parser --search 12  --search_alpha 1e-4 --search_rollout oracle --holdout_off
     train-sets/ref/search_dep_parser.stderr
 
 # Test 67: classification with data from dictionaries (eg embeddings or gazetteers) -- note that this is impossible without dictionaries because --ignore w
diff --git a/test/train-sets/ref/search_dep_parser.stderr b/test/train-sets/ref/search_dep_parser.stderr
index 9bb34df607d..7e68d31c44d 100644
--- a/test/train-sets/ref/search_dep_parser.stderr
+++ b/test/train-sets/ref/search_dep_parser.stderr
@@ -9,14 +9,15 @@ num sources = 1
 average    since      instance            current true      current predicted   cur   cur   predic    cache  examples          
 loss       last        counter           output prefix          output prefix  pass   pol     made     hits    gener  beta    
 88.000000  88.000000         1  [43:1 5:2 5:2 5:2 1..] [0:8 1:1 2:1 3:1 4:..]     0     0      144        0      144  0.014199
-47.500000  7.000000          2  [2:2 3:5 0:8 3:7 3:4 ] [2:2 0:8 2:5 2:3 2:4 ]     0     0      157        0      156  0.015381
+47.500000  7.000000          2  [2:2 3:5 0:8 3:7 3:4 ] [2:2 5:2 2:4 2:4 0:8 ]     0     0      157        0      156  0.015381
 38.250000  29.000000         4  [4:2 4:2 4:2 7:5 6:..] [2:2 0:8 2:4 2:1 4:..]     0     0      248        0      246  0.024204
-28.125000  18.000000         8  [4:2 4:2 4:2 5:5 0:..] [3:2 3:2 4:2 5:5 0:..]     1     0      551        0      543  0.052760
+29.375000  20.500000         8  [4:2 4:2 4:2 5:5 0:..] [4:2 3:2 4:2 5:5 0:..]     1     0      551        0      543  0.052760
+19.812500  10.250000        16  [43:1 5:2 5:2 5:2 1..] [30:1 5:2 5:2 5:2 1..]     3     0     1187        0     1134  0.107122
 
 finished run
 number of examples per pass = 5
-passes used = 2
-weighted example sum = 10
+passes used = 6
+weighted example sum = 30
 weighted label sum = 0
-average loss = 23.9
-total feature number = 275880
+average loss = 10.5667
+total feature number = 827640