How do I remove duplicate records in a join table in PostgreSQL?

Question

I have a table that has a schema like this:

create_table "questions_tags", :id => false, :force => true do |t|
        t.integer "question_id"
        t.integer "tag_id"
      end

      add_index "questions_tags", ["question_id"], :name => "index_questions_tags_on_question_id"
      add_index "questions_tags", ["tag_id"], :name => "index_questions_tags_on_tag_id"

I would like to remove records that are duplicates, i.e. they have both the same tag_id and question_id as another record.

What does the SQL look like for that?