diff --git a/src/pie_datasets/document/types.py b/src/pie_datasets/document/types.py
deleted file mode 100644
index 0ea32ed5..00000000
--- a/src/pie_datasets/document/types.py
+++ /dev/null
@@ -1,15 +0,0 @@
-import dataclasses
-
-from pytorch_ie.annotations import BinaryRelation, LabeledSpan
-from pytorch_ie.core import AnnotationList, annotation_field
-from pytorch_ie.documents import TokenBasedDocument
-
-
-@dataclasses.dataclass
-class TokenDocumentWithLabeledSpans(TokenBasedDocument):
-    labeled_spans: AnnotationList[LabeledSpan] = annotation_field(target="tokens")
-
-
-@dataclasses.dataclass
-class TokenDocumentWithLabeledSpansAndBinaryRelations(TokenDocumentWithLabeledSpans):
-    binary_relations: AnnotationList[BinaryRelation] = annotation_field(target="labeled_spans")
diff --git a/tests/dataset_builders/common.py b/tests/dataset_builders/common.py
index 70af75a7..e04169c4 100644
--- a/tests/dataset_builders/common.py
+++ b/tests/dataset_builders/common.py
@@ -1,3 +1,4 @@
+import dataclasses
 import json
 import logging
 import os
@@ -5,6 +6,10 @@
 from pathlib import Path
 from typing import List, Optional
 
+from pytorch_ie.annotations import BinaryRelation, LabeledSpan
+from pytorch_ie.core import AnnotationList, annotation_field
+from pytorch_ie.documents import TokenBasedDocument
+
 from tests import FIXTURES_ROOT
 
 DATASET_BUILDER_BASE_PATH = Path("dataset_builders")
@@ -68,3 +73,13 @@ def _load_json(fn: str):
     with open(fn) as f:
         ex = json.load(f)
     return ex
+
+
+@dataclasses.dataclass
+class TestTokenDocumentWithLabeledSpans(TokenBasedDocument):
+    labeled_spans: AnnotationList[LabeledSpan] = annotation_field(target="tokens")
+
+
+@dataclasses.dataclass
+class TestTokenDocumentWithLabeledSpansAndBinaryRelations(TestTokenDocumentWithLabeledSpans):
+    binary_relations: AnnotationList[BinaryRelation] = annotation_field(target="labeled_spans")
diff --git a/tests/dataset_builders/pie/test_cdcp.py b/tests/dataset_builders/pie/test_cdcp.py
index 86e6d3c1..643d7b6f 100644
--- a/tests/dataset_builders/pie/test_cdcp.py
+++ b/tests/dataset_builders/pie/test_cdcp.py
@@ -20,9 +20,12 @@
     example_to_document,
 )
 from pie_datasets import DatasetDict
-from pie_datasets.document.types import TokenDocumentWithLabeledSpansAndBinaryRelations
 from tests import FIXTURES_ROOT
-from tests.dataset_builders.common import PIE_BASE_PATH, _deep_compare
+from tests.dataset_builders.common import (
+    PIE_BASE_PATH,
+    TestTokenDocumentWithLabeledSpansAndBinaryRelations,
+    _deep_compare,
+)
 
 disable_caching()
 
@@ -306,7 +309,7 @@ def tokenizer() -> PreTrainedTokenizer:
 @pytest.fixture(scope="module")
 def tokenized_documents_with_labeled_spans_and_binary_relations(
     dataset_of_text_documents_with_labeled_spans_and_binary_relations, tokenizer
-) -> List[TokenDocumentWithLabeledSpansAndBinaryRelations]:
+) -> List[TestTokenDocumentWithLabeledSpansAndBinaryRelations]:
     # get a document to check
     doc = dataset_of_text_documents_with_labeled_spans_and_binary_relations["train"][0]
     # Note, that this is a list of documents, because the document may be split into chunks
@@ -315,7 +318,7 @@ def tokenized_documents_with_labeled_spans_and_binary_relations(
         doc,
         tokenizer=tokenizer,
         return_overflowing_tokens=True,
-        result_document_type=TokenDocumentWithLabeledSpansAndBinaryRelations,
+        result_document_type=TestTokenDocumentWithLabeledSpansAndBinaryRelations,
         verbose=True,
     )
     return tokenized_docs
@@ -433,7 +436,7 @@ def test_tokenized_documents_with_entities_and_relations_all(
                 doc,
                 tokenizer=tokenizer,
                 return_overflowing_tokens=True,
-                result_document_type=TokenDocumentWithLabeledSpansAndBinaryRelations,
+                result_document_type=TestTokenDocumentWithLabeledSpansAndBinaryRelations,
                 verbose=True,
             )
             # we just ensure that we get at least one tokenized document
diff --git a/tests/dataset_builders/pie/test_scidtb_argmin.py b/tests/dataset_builders/pie/test_scidtb_argmin.py
index b7bced97..f26eda63 100644
--- a/tests/dataset_builders/pie/test_scidtb_argmin.py
+++ b/tests/dataset_builders/pie/test_scidtb_argmin.py
@@ -16,9 +16,12 @@
     example_to_document,
 )
 from pie_datasets import DatasetDict
-from pie_datasets.document.types import TokenDocumentWithLabeledSpansAndBinaryRelations
 from tests import FIXTURES_ROOT
-from tests.dataset_builders.common import HF_DS_FIXTURE_DATA_PATH, PIE_BASE_PATH
+from tests.dataset_builders.common import (
+    HF_DS_FIXTURE_DATA_PATH,
+    PIE_BASE_PATH,
+    TestTokenDocumentWithLabeledSpansAndBinaryRelations,
+)
 
 disable_caching()
 
@@ -292,7 +295,7 @@ def tokenizer() -> PreTrainedTokenizer:
 @pytest.fixture(scope="module")
 def tokenized_documents_with_labeled_spans_and_binary_relations(
     dataset_of_text_documents_with_labeled_spans_and_binary_relations, tokenizer
-) -> List[TokenDocumentWithLabeledSpansAndBinaryRelations]:
+) -> List[TestTokenDocumentWithLabeledSpansAndBinaryRelations]:
     # get a document to check
     doc = dataset_of_text_documents_with_labeled_spans_and_binary_relations["train"][0]
     # Note, that this is a list of documents, because the document may be split into chunks
@@ -301,7 +304,7 @@ def tokenized_documents_with_labeled_spans_and_binary_relations(
         doc,
         tokenizer=tokenizer,
         return_overflowing_tokens=True,
-        result_document_type=TokenDocumentWithLabeledSpansAndBinaryRelations,
+        result_document_type=TestTokenDocumentWithLabeledSpansAndBinaryRelations,
         verbose=True,
     )
     return tokenized_docs
@@ -373,7 +376,7 @@ def test_tokenized_documents_with_entities_and_relations_all(
                 doc,
                 tokenizer=tokenizer,
                 return_overflowing_tokens=True,
-                result_document_type=TokenDocumentWithLabeledSpansAndBinaryRelations,
+                result_document_type=TestTokenDocumentWithLabeledSpansAndBinaryRelations,
                 verbose=True,
             )
             # we just ensure that we get at least one tokenized document