[2607.17535] Salience Induction against Multi-Hop RAG Agents: Threat and Defense

Skip to main content

Search arXiv

Press Enter to search · Advanced search

-->

Computer Science > Cryptography and Security

arXiv:2607.17535 (cs)

[Submitted on 20 Jul 2026]

Title:Salience Induction against Multi-Hop RAG Agents: Threat and Defense

Authors:Xingfu Zhou, Pengfei Wang, Yuan Zhou, Wei Xie, Xu Zhou<br>View a PDF of the paper titled Salience Induction against Multi-Hop RAG Agents: Threat and Defense, by Xingfu Zhou and 4 other authors

View PDF<br>HTML (experimental)

Abstract:Agentic retrieval-augmented generation (RAG) systems increasingly retrieve external evidence and orchestrate tools for knowledge-intensive applications. In Multi-Hop question answering, agents chain facts across documents. Existing defenses focus on content poisoning, which injects false facts, and prompt injection, which embeds directives. We identify a third attack surface: the salience channel, through which fact position, emphasis, framing, and semantic proximity can redirect reasoning even when all retrieved claims are true and no instructions are present. We formalize Salience Induction as truth-preserving edits that redirect Multi-Hop attribute binding while leaving the retrieval trace semantically intact. We define six Salience-Editing operator classes and build an iterative proposer-verifier pipeline under factual and stealth constraints. We also introduce SalientWiki-MH, a decoy-annotated Multi-Hop benchmark. Evaluations across five frontier model families (GPT, Claude, Gemini, DeepSeek, and Qwen) and three agent architectures (ReAct, Reflexion, and tool-calling) show broad generalization. Under a 30% edit budget, Salience Induction achieves an 83.3% attack success rate; the strongest evaluated baseline defense leaves 75.7% post-defense ASR. Untargeted rewriting further reduces attacks only by degrading neutral task success. Our lightweight input-side defense, Salience Normalization, reduces attack success to 15.3% under standard attacks and 23.6% under an adaptive attack. These results show that truthfulness and instruction filtering alone are insufficient: robust agentic RAG also requires defenses against salience-relevance decoupling.

Comments:<br>18 pages, 4 figures, 12 tables

Subjects:

Cryptography and Security (cs.CR); Computation and Language (cs.CL)

Cite as:<br>arXiv:2607.17535 [cs.CR]

(or<br>arXiv:2607.17535v1 [cs.CR] for this version)

https://doi.org/10.48550/arXiv.2607.17535

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

Submission history<br>From: Xingfu Zhou [view email]<br>[v1]<br>Mon, 20 Jul 2026 04:27:50 UTC (1,787 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled Salience Induction against Multi-Hop RAG Agents: Threat and Defense, by Xingfu Zhou and 4 other authors<br>View PDF<br>HTML (experimental)<br>TeX Source

view license

Current browse context:

cs.CR

next >

new<br>recent<br>| 2026-07

Change to browse by:

cs<br>cs.CL

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

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 .

Which authors of this paper are endorsers? |<br>Disable MathJax (What is MathJax?)

Major funding support from