Metraipolis › archive › dc_3dd04ba055ee

Beyond Similarity through Zero-Token Geometric Graphs for Multi-Hop RAG — arXiv:2609.19622

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.19622
License (as archived): CC0-1.0
Archived: 2026-10-10T05:05:16.630Z
SHA-256 of the archived text: 2b532ad9df1dfe1586593193a9169ab0419f56f602f2dcb867ca937cc1f1bdbb
Stable URL of this page: https://metraipolis.com/documento/dc_3dd04ba055ee

How to cite

Beyond Similarity through Zero-Token Geometric Graphs for Multi-Hop RAG — arXiv:2609.19622 — arXiv (en), https://arxiv.org/abs/2609.19622 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3dd04ba055ee (Metraipolis archive dc_3dd04ba055ee, sha256 2b532ad9df1dfe1586593193a9169ab0419f56f602f2dcb867ca937cc1f1bdbb)

BibTeX

@misc{dc3dd04ba055ee,
  title = {Beyond Similarity through Zero-Token Geometric Graphs for Multi-Hop RAG — arXiv:2609.19622},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19622},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3dd04ba055ee; sha256 2b532ad9df1dfe1586593193a9169ab0419f56f602f2dcb867ca937cc1f1bdbb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19622. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-hop retrieval-augmented generation (RAG) requires evidence that remains relevant to a query while introducing enough novelty to bridge semantic gaps. Dense retrieval tends to concentrate on semantically similar documents, whereas graph-based alternatives often depend on costly Large Language Model (LLM) entity extraction and may propagate through noisy connections. We introduce Geometric Gain Graph RAG (G$^3$RAG), a document-only framework whose offline graph construction uses no LLM calls or generated tokens. G$^3$RAG assigns each edge a geometric gain score, $\cos\theta \cdot \sin\theta$, that jointly captures directional consistency and orthogonality between document representations. A density-aware topological penalty suppresses highly connected hubs, while single-step controlled diffusion expands from filtered query seeds toward complementary evidence. We evaluate G$^3$RAG on MusiQue, 2WikiMultiHopQA, and HotpotQA using Nv-embed-v2 and Qwen3-8B-embed. G$^3$RAG obtains the best average F1 and answer-document hit rate among the evaluated graph-based baselines in both embedding settings, with gains of up to 4.26 F1 points in average performance and 5.76 points on MusiQue. It also removes the graph-construction token cost incurred by entity-based graph methods. These results show that geometric structure can support efficient multi-hop evidence discovery without LLM-based graph construction. Code is available at this https URL

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-18 in cs.IR).
Categorie arXiv: Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2609.19622 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T04:24:02.352Z dalla pagina https://arxiv.org/catchup/cs.IR/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).