PluriHop: Exhaustive, Recall-Sensitive QA over Distractor-Rich Corpora
This addresses a practical challenge for domains like medical records and compliance filings where missing even one document can lead to incorrect answers, though it is incremental in improving retrieval-augmented generation for specific data types.
The paper tackles the problem of question answering over repetitive, distractor-rich corpora where retrieval must be exhaustive and recall-sensitive, introducing a new dataset and method that achieves relative F1 score improvements of 18-52% over baselines.
Recent advances in large language models (LLMs) and retrieval-augmented generation (RAG) have enabled progress on question answering (QA) when relevant evidence is in one (single-hop) or multiple (multi-hop) passages. Yet many realistic questions about recurring report data - medical records, compliance filings, maintenance logs - require aggregation across all documents, with no clear stopping point for retrieval and high sensitivity to even one missed passage. We term these pluri-hop questions and formalize them by three criteria: recall sensitivity, exhaustiveness, and exactness. To study this setting, we introduce PluriHopWIND, a diagnostic multilingual dataset of 48 pluri-hop questions built from 191 real-world wind industry reports in German and English. We show that PluriHopWIND is 8-40% more repetitive than other common datasets and thus has higher density of distractor documents, better reflecting practical challenges of recurring report corpora. We test a traditional RAG pipeline as well as graph-based and multimodal variants, and find that none of the tested approaches exceed 40% in statement-wise F1 score. Motivated by this, we propose PluriHopRAG, a RAG architecture that follows a "check all documents individually, filter cheaply" approach: it (i) decomposes queries into document-level subquestions and (ii) uses a cross-encoder filter to discard irrelevant documents before costly LLM reasoning. We find that PluriHopRAG achieves relative F1 score improvements of 18-52% depending on base LLM. Despite its modest size, PluriHopWIND exposes the limitations of current QA systems on repetitive, distractor-rich corpora. PluriHopRAG's performance highlights the value of exhaustive retrieval and early filtering as a powerful alternative to top-k methods.