A retrieval benchmark over Indonesian legislation from BPK JDIH (peraturan.bpk.go.id). Built for an undergraduate thesis at the Faculty of Computer Science, Universitas Indonesia, that compares a BM25 lexical baseline, vectorless retrieval driven by LLM reasoning, and vector-based dense retrieval on the same corpus and gold set. The benchmark is retrieval-only. There is no answer generation and no generation labels, systems are scored on… See the full description on the dataset page:
https://huggingface.co/datasets/wahyyuht/skripsi-data.
