InverseMapper causes IntWritable type to collected as Text

When using the org.apache.hadoop.mapred.lib.InverseMapper, key,values are inversed as expected. For example (Text,IntWritable) will get inversed. However, output key,value only works if you use (Text, Text). Below is an example, where I was chaining 2 jobs:

Job 1:
public static class Reduce extends MapReduceBase implements Reducer<Text, IntWritable, Text, IntWritable>

{...}

Job 2:
conf.setMapperClass(InverseMapper.class);
public static class Reduce extends MapReduceBase implements Reducer<IntWritable, Text, Text, IntWritable> {...}

//I would expect this to work. When I do this, I get
java.lang.ClassCastException: org.apache.hadoop.io.Text cannot be cast to org.apache.hadoop.io.IntWritable

To re-inverse my key,values, I had to do this:
public static class Reduce extends MapReduceBase implements Reducer<Text, Text, Text, Text>

Notice that in order for the reducer to properly accept key,values, I had to indicate the key as Text.