BridgeRAG: Training-Free Bridge-Conditioned Retrieval for Multi-Hop Question Answering

• URL: http://arxiv.org/abs/2604.03384v1
• Date: Fri, 03 Apr 2026 18:27:51 GMT
• Title: BridgeRAG: Training-Free Bridge-Conditioned Retrieval for Multi-Hop Question Answering
• Abstract summary: We present BridgeRAG, a training-free, graph-free retrieval method for retrieval-augmented generation.<n>BridgeRAG separates coverage from scoring: dual-entity ANN broadens the second-hop candidate pool.<n>We show that bridge conditioning causes productive re-rankings (18.7% flip-win rate on parallel-chain vs. 0.6% on single-chain)
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-hop retrieval is not a single-step relevance problem: later-hop evidence should be ranked by its utility conditioned on retrieved bridge evidence, not by similarity to the original query alone. We present BridgeRAG, a training-free, graph-free retrieval method for retrieval-augmented generation (RAG) over multi-hop questions that operationalizes this view with a tripartite scorer s(q,b,c) over (question, bridge, candidate). BridgeRAG separates coverage from scoring: dual-entity ANN expansion broadens the second-hop candidate pool, while a bridge-conditioned LLM judge identifies the active reasoning chain among competing candidates without any offline graph or proposition index. Across four controlled experiments we show that this conditioning signal is (i) selective: +2.55pp on parallel-chain queries (p<0.001) vs. ~0 on single-chain subtypes; (ii) irreplaceable: substituting the retrieved passage with generated SVO query text reduces R@5 by 2.1pp, performing worse than even the lowest-SVO-similarity pool passage; (iii) predictable: cos(b,g2) correlates with per-query gain (Spearman rho=0.104, p<0.001); and (iv) mechanistically precise: bridge conditioning causes productive re-rankings (18.7% flip-win rate on parallel-chain vs. 0.6% on single-chain), not merely more churn. Combined with lightweight coverage expansion and percentile-rank score fusion, BridgeRAG achieves the best published training-free R@5 under matched benchmark evaluation on all three standard MHQA benchmarks without a graph database or any training: 0.8146 on MuSiQue (+3.1pp vs. PropRAG, +6.8pp vs. HippoRAG2), 0.9527 on 2WikiMultiHopQA (+1.2pp vs. PropRAG), and 0.9875 on HotpotQA (+1.35pp vs. PropRAG).

Related papers

• Dissecting Agentic RAG: A Component Ablation for Multi-Hop QA with a Local 7B Model [0.0]
  We run a controlled ablation study of a full agentic RAG pipeline evaluated on 5,000 questions.<n>We find that fixed hybrid retrieval via reciprocal rank fusion consistently outperforms rule-based adaptive routing.
  arXiv  Detail & Related papers  (2026-06-19T15:50:35Z)
• SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale [54.70985426016736]
  We present SkillDAG, which models inter-skill relationships as a typed directed graph.<n>Each search returns vector matches, typed-edge neighbors, and conflict signals, and a propose-then-commit protocol lets the agent register execution-backed edges.<n>On ALFWorld and SkillsBench with MiniMax-M2.7, SkillDAG reaches 67.1% success and 27.3% reward.
  arXiv  Detail & Related papers  (2026-06-02T02:45:21Z)
• PAR$^2$-RAG: Planned Active Retrieval and Reasoning for Multi-Hop Question Answering [57.89576196160413]
  Large language models (LLMs) remain brittle on multi-hop question answering (MHQA)<n>We propose textbfPlanned Active Retrieval and Reasoning RAG (PAR$2$-RAG), a two-stage framework that separates emphcoverage from emphcommitment.
  arXiv  Detail & Related papers  (2026-03-30T23:52:54Z)
• Calibrated Fusion for Heterogeneous Graph-Vector Retrieval in Multi-Hop QA [0.0]
  We map vector and graph scores to a common unit-free scale using percentile-rank normalization (PIT) before fusion.<n>Across MuSiQue and 2WikiMultiHopQA, calibrated fusion improves held-out last-hop retrieval.
  arXiv  Detail & Related papers  (2026-03-30T18:13:01Z)
• The Reasoning Bottleneck in Graph-RAG: Structured Prompting and Context Compression for Multi-Hop QA [4.190399245544758]
  Graph-RAG systems achieve strong multi-hop question answering by indexing documents into knowledge graphs, but strong retrieval does not guarantee strong answers.<n>We show that 77% to 91% of questions have the gold answer in the retrieved context, yet accuracy is only 35% to 78%, and 73% to 84% of errors are reasoning failures.<n>We propose two augmentations: (i) SPARQL chain-of-thought prompting, which decomposes questions into triple-pattern queries aligned with the entity-relationship context, and (ii) graph-walk compression.
  arXiv  Detail & Related papers  (2026-03-14T17:29:13Z)
• StratRAG: A Multi-Hop Retrieval Evaluation Dataset for Retrieval-Augmented Generation Systems [0.0]
  StratRAG is an evaluation dataset for benchmarking Retrieval-Augmented Generation (RAG) systems.<n>It comprises 2,200 examples across three question types -- bridge, comparison, and yes-no -- each paired with a pool of 15 candidate documents.<n>We benchmark three retrieval strategies -- BM25, dense retrieval, and hybrid fusion.
  arXiv  Detail & Related papers  (2026-03-06T20:08:06Z)
• P-RAG: Prompt-Enhanced Parametric RAG with LoRA and Selective CoT for Biomedical and Multi-Hop QA [9.399056753263757]
  Retrieval-Augmented Generation (RAG) addresses this constraint by retrieving external knowledge during inference.<n>We evaluate three RAG variants-Standard RAG, DA-RAG, and our proposed Prompt-Enhanced Parametric RAG (P-RAG)<n>P-RAG integrates parametric knowledge within the LLM and retrieved evidence, guided by Chain-of-Thought (CoT) prompting and Low-Rank Adaptation (LoRA)
  arXiv  Detail & Related papers  (2026-02-02T03:42:45Z)
• CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal [84.71254539482369]
  Group-relative reinforcement learning with verifiable rewards (RLVR) often wastes the most informative data it already has the failures.<n>We present CARE, a failure-centric post-training framework for multimodal reasoning that turns errors into supervision.<n> CARE improves accuracy and training smoothness while explicitly increasing the share of learning signal that comes from failures.
  arXiv  Detail & Related papers  (2025-12-22T16:34:21Z)
• Generalized Parallel Scaling with Interdependent Generations [58.43994876504917]
  We propose Bridge to generate interdependent responses in parallel.<n>With only a small amount of new parameters, Bridge improves the relative mean accuracy gains from reinforcement learning.
  arXiv  Detail & Related papers  (2025-10-01T17:33:35Z)
• PropRAG: Guiding Retrieval with Beam Search over Proposition Paths [15.346744525284604]
  PropRAG is a novel RAG framework that shifts from triples to context-rich propositions.<n>PropRAG achieves Recall@5 and F1 scores on 2Wiki, HotpotQA, and MuSiQue.
  arXiv  Detail & Related papers  (2025-04-25T04:47:34Z)
• Hydra-SGG: Hybrid Relation Assignment for One-stage Scene Graph Generation [57.69385990442078]
  Hydra-SGG achieves state-of-the-art performance on multiple datasets, including VG150 (16.0 mR@50), Open Images V6 (50.1 weighted score), and GQA (12.7 mR@50)
  arXiv  Detail & Related papers  (2024-09-16T13:13:06Z)
• Faithful Chain-of-Thought Reasoning [51.21714389639417]
  Chain-of-Thought (CoT) prompting boosts Language Models' (LM) performance on a gamut of reasoning tasks. We propose Faithful CoT, a reasoning framework involving two stages: Translation and Problem Solving. This guarantees that the reasoning chain provides a faithful explanation of the final answer.
  arXiv  Detail & Related papers  (2023-01-31T03:04:26Z)
• Adversarial Retriever-Ranker for dense text retrieval [51.87158529880056]
  We present Adversarial Retriever-Ranker (AR2), which consists of a dual-encoder retriever plus a cross-encoder ranker. AR2 consistently and significantly outperforms existing dense retriever methods. This includes the improvements on Natural Questions R@5 to 77.9%(+2.1%), TriviaQA R@5 to 78.2%(+1.4), and MS-MARCO MRR@10 to 39.5%(+1.3%)
  arXiv  Detail & Related papers  (2021-10-07T16:41:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.