Faithful or Findable? Evaluating LLM-Generated Metadata for RDF Dataset Search
For dataset search systems, it highlights the need to jointly evaluate retrieval effectiveness and faithfulness of synthetic metadata, showing that gains can come at the cost of trustworthiness.
The paper evaluates LLM-generated metadata for RDF dataset search, finding that unconstrained rewriting improves retrieval but reduces faithfulness, while profile-grounded rewriting offers the best trade-off between retrieval effectiveness and grounding.
Dataset search depends heavily on metadata, making LLM-generated metadata a consequential form of synthetic content in retrieval systems. We study six metadata-generation settings for RDF datasets, ranging from simple rewriting to profile-grounded and agentic graph-based generation, and evaluate them jointly for retrieval effectiveness and faithfulness. Unconstrained metadata rewriting delivers the strongest retrieval gains over the original metadata, but it is also the least faithful, showing that search improvements can be driven by unsupported semantic expansion. More grounded settings substantially improve faithfulness, and profile-grounded rewriting provides the most balanced trade-off between retrieval effectiveness and grounding. These findings position synthetic metadata as a system-level IR problem in which effectiveness, provenance, and trust must be evaluated together.