The schema output is different (displayed via the getSchema method of the DATASET type).
SELECT data.getSchema() FROM DATASET_PUBLISH ( ON (SELECT * FROM employeeTable) USING UNIQUE_NAMES('Y') ) AS avroFiles;
Result:
data.getSchema() { "type": "array", "items": { "type": "record", "name": "rec_0_1448383894", "fields": [{ "name": "empID", "type": "int" }, { "name": "empName", "type": "string" }, { "name": "empDept", "type": "string" }] } }