diff --git a/src/xpmir/conversation/learning/__init__.py b/src/xpmir/conversation/learning/__init__.py
index 23eee83..4b0dc17 100644
--- a/src/xpmir/conversation/learning/__init__.py
+++ b/src/xpmir/conversation/learning/__init__.py
@@ -1,4 +1,5 @@
 from functools import cached_property
+from datamaestro.record import RecordTypesCache
 import numpy as np
 from datamaestro_text.data.ir import TopicRecord
 from datamaestro_text.data.conversation import (
@@ -22,6 +23,11 @@ class DatasetConversationEntrySampler(BaseSampler):
     def conversations(self):
         return list(self.dataset.__iter__())
 
+    def __post_init__(self):
+        super().__post_init__()
+
+        self._recordtypes = RecordTypesCache()
+
     def __iter__(self) -> RandomSerializableIterator[TopicConversationRecord]:
         def generator(random: np.random.RandomState):
             while True:
@@ -38,8 +44,8 @@ def generator(random: np.random.RandomState):
                 node_ix = random.randint(len(nodes))
                 node = nodes[node_ix]
 
-                node = node.entry().add(
-                    ConversationHistoryItem(node.history()), no_check=True
+                node = self._recordtypes.update(
+                    node.entry(), ConversationHistoryItem(node.history())
                 )
 
                 yield node
diff --git a/src/xpmir/index/sparse.py b/src/xpmir/index/sparse.py
index 18a6daf..7e7589b 100644
--- a/src/xpmir/index/sparse.py
+++ b/src/xpmir/index/sparse.py
@@ -236,6 +236,7 @@ def execute(self):
         ]
 
         # Cleanup the index before starting
+        # ENHANCE: recover index build when possible
         from shutil import rmtree
 
         if self.index_path.is_dir():
diff --git a/src/xpmir/interfaces/anserini.py b/src/xpmir/interfaces/anserini.py
index 9710027..45176ea 100644
--- a/src/xpmir/interfaces/anserini.py
+++ b/src/xpmir/interfaces/anserini.py
@@ -12,7 +12,7 @@
 from typing import List, Optional
 from experimaestro import tqdm as xpmtqdm, Task, Meta
 
-from datamaestro_text.data.ir import DocumentStore, TextItem, IDItem
+from datamaestro_text.data.ir import DocumentStore, TextItem, IDItem, TopicRecord
 import datamaestro_text.data.ir.csv as ir_csv
 from datamaestro_text.data.ir.trec import (
     Documents,
@@ -273,7 +273,7 @@ class AnseriniRetriever(Retriever):
 
     Attributes:
         index: The Anserini index
-        model: the model used to search. Only suupports BM25 so far.
+        model: the model used to search. Only supports BM25 so far.
         k: Number of results to retrieve
     """
 
@@ -306,10 +306,10 @@ def _get_store(self) -> Optional[Index]:
         if self.index.storeContents:
             return self.index
 
-    def retrieve(self, query: str) -> List[ScoredDocument]:
+    def retrieve(self, record: TopicRecord) -> List[ScoredDocument]:
         # see
         # https://github.com/castorini/anserini/blob/master/src/main/java/io/anserini/search/SimpleSearcher.java
-        hits = self.searcher.search(query, k=self.k)
+        hits = self.searcher.search(record[TextItem].text, k=self.k)
         store = self.get_store()
 
         # Batch retrieve documents
diff --git a/src/xpmir/rankers/full.py b/src/xpmir/rankers/full.py
index 2027fcf..2b70a37 100644
--- a/src/xpmir/rankers/full.py
+++ b/src/xpmir/rankers/full.py
@@ -20,7 +20,7 @@ class FullRetriever(Retriever):
 
     documents: Param[Documents]
 
-    def retrieve(self, query: str) -> List[ScoredDocument]:
+    def retrieve(self, record: TopicRecord) -> List[ScoredDocument]:
         return [ScoredDocument(doc, 0.0) for doc in self.documents]
 
 
@@ -111,9 +111,9 @@ def score(
         # Add each result to the full document list
         scored_documents.extend(new_scores)
 
-    def retrieve(self, query: str):
+    def retrieve(self, record: TopicRecord) -> List[ScoredDocument]:
         # Only use retrieve_all
-        return self.retrieve_all({"_": query})["_"]
+        return self.retrieve_all({"_": record})["_"]
 
     def retrieve_all(
         self, queries: Dict[str, TopicRecord]