aiqbench analysis
This paper addresses a critical failure mode in Retrieval-Augmented Generation (RAG) where standard similarity metrics retrieve documents based on keyword association rather than causal relevance to the query. By modeling the retrieval process itself through a causal graph grounded in Reichenbach's common cause principle, the authors introduce a training-free attention mechanism that significantly improves precision in enterprise knowledge bases prone to keyword-stuffing attacks. While performance drops on general BEIR benchmarks, the method offers a robust, deployable solution for specific regimes where traditional rerankers fail to distinguish between associative noise and causal information.
From Association to Causation: Improving Retrieval Precision of Retrieval-Augmented Generation via Causal Relations and an Attention Mechanism
arXiv:2608.21702v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) grounds LLM generation on retrieved documents, but the standard terminal retrieval stage--dense-vector similarity, optionally followed by reranking--often returns documents that share keywords with the query without containing the needed information, a failure mode that grows with the knowledge base. We trace it to a conceptual gap: similarity captures only associational relations, whereas the documents that matter are linked to the query causally. We model the terminal retrieval stage with a causal graph grounded in Reichenbach's common cause principle: the keywords shared by the query and a retrieved document form a latent common cause A, and the document's residual keywords form a latent set B linking the document to the ideal output. Since a retrieved document is a collider (A -> d = 95% reliability. A fully local testbed demonstrates deployability.
Abstract:Retrieval-Augmented Generation (RAG) grounds LLM generation on retrieved documents, but the standard terminal retrieval stage--dense-vector similarity, optionally followed by reranking--often returns documents that share keywords with the query without containing the needed information, a failure mode that grows with the knowledge base. We trace it to a conceptual gap: similarity captures only associational relations, whereas the documents that matter are linked to the query causally. We model the terminal retrieval stage with a causal graph grounded in Reichenbach's common cause principle: the keywords shared by the query and a retrieved document form a latent common cause A, and the document's residual keywords form a latent set B linking the document to the ideal output. Since a retrieved document is a collider (A -> d <- B), retrieval itself opens an associational path between the query and B, which licenses a training-free, attention-style re-scoring rule: the cosine similarity between the query embedding and the weighted centroid embedding of B. Unlike causality-enhanced RAG variants that model causal relations inside the knowledge content, our graph models the causal structure of the retrieval process itself. On a real 471-document enterprise knowledge base, the method promotes a relevant guideline from rank 6 to the top 3; on a controlled diagnostic corpus reproducing the keyword-stuffing regime, it improves the mean target rank from 2.88 to 1.25, while a trained cross-encoder reranker barely helps (2.63). Conversely, on three BEIR benchmarks the score underperforms the similarity baseline, delineating the applicability boundary: the method guards the keyword-stuffing regime of growing proprietary knowledge bases and complements neural rerankers; a corpus-level calibration gate selects the correct regime with >= 95% reliability. A fully local testbed demonstrates deployability.
16 pages, 2 figures, 4 tables, 1 algorithm. Code available at this https URL
Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
arXiv:2608.21702 [cs.AI]
(or arXiv:2608.21702v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2608.21702 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 388 words
Related coverage
The Effect of Text Chunk Size on Retrieval-Augmented Generation Performance
arXiv · Jul 29, 2026
GuidedRAG: Semantic Steering of Retrieval-Augmented Generation
arXiv · Jul 31, 2026
Retrieval-Augmented Generation in LLMs for Mental Health: Quantifying the Incremental Contribution of Retrieval Within a Layered Safety Architecture
arXiv · Jul 29, 2026
Crash Narrative-Guided Countermeasure Recommendation Using Large Language Models: A Retrieval-Augmented Generation Framework for Intersection Safety
arXiv · Sep 16, 2026