paper-with-me

홈 › Papers

Regime-Conditional Retrieval: Theory and a Transferable Router for Two-Hop QA

2026-04-10 · Andre Bacellar arxiv

Two-hop QA retrieval splits queries into two regimes determined by whether the hop-2 entity is explicitly named in the question (Q-dominant) or only in the bridge passage (B-dominant). We formalize this split with three theorems: (T1) per-query AUC is a monotone function of the cosine separation margin, with R^2 >= 0.90 for six of eight type-encoder pairs; (T2) regime is characterized by two surface-text predicates, with P1 decisive for routing and P2 qualifying the B-dominant case, holding across three encoders and three datasets; and (T3) bridge advantage requires the relation-bearing sentence, not entity name alone, with removal causing an 8.6-14.1 pp performance drop (p < 0.001). Building on this theory, we propose RegimeRouter, a lightweight binary router that selects between question-only and question-plus-relation-sentence retrieval using five text features derived directly from the predicate definitions. Trained on 2WikiMultiHopQA (n = 881, 5-fold cross-fitted) and applied zero-shot to MuSiQue and HotpotQA, RegimeRouter achieves +5.6 pp (p < 0.001), +5.3 pp (p = 0.002), and +1.1 pp (non-significant, no-regret) R@5 improvement, respectively, with artifact-driven.

📄 PDF Abstract BibTeX arXiv:2604.09019

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

A Regime Theory of Controller Class Selection for LLM Action Decisions

2026-05-07 · Zhaoyang Jiang, Zhizhong Fu, Yunsoo Kim, Jiacong Mi 외 arxiv

Deployed language and vision-language models must decide, on each input, whether to answer directly, retrieve evidence, defer to a stronger model, or abstain. Contrary to the common monotonicity intuition, greater per-in…

TriForces: Augmenting Atomistic GNNs for Transferable Representations

2026-05-20 · Ali Ramlaoui, Alexandre Duval, Hannah Bull, Victor Schmidt 외 arxiv

Machine learning interatomic potentials (MLIPs) achieve excellent accuracy when trained on large Density Functional Theory (DFT) data. To be useful in practice, they must often be adapted to target chemistries using smal…

Self-Supervised Learning

Trust Before Fusion: QIMG-7 and Source-Aware Resolution for Polluted Multimodal RAG

2026-07-12 · Saadeldine Eletter, Owais Aijaz, Preslav Nakov arxiv

Multimodal retrieval-augmented generation (RAG) is often evaluated with clean evidence, yet real retrieval can return topically relevant but unreliable content: false text and misleading images from corrupted metadata, e…

Style Transfer

RetrievalRouter: Joint Modality and Architecture Selection for Document Retrieval

2026-08-26 · Emre Kuru, Mehmet Onur Keskin, Reza Farahbakhsh, Noel Crespi hf

Document retrieval increasingly supports high-stakes information access in finance, healthcare, and law. Modern retrieval pipelines vary both in modality (text or multimodal) and in retrieval architecture (dense or late-…

Lightweight Query Routing for Adaptive RAG: A Baseline Study on RAGRouter-Bench

2026-04-03 · Prakhar Bansal, Shivangi Agarwal arxiv

Retrieval-Augmented Generation pipelines span a wide range of retrieval strategies that differ substantially in token cost and capability. Selecting the right strategy per query is a practical efficiency problem, yet no …