PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing

PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing | Signal Canvas | ScienceToStartup

Page Freshness

Signal Canvas proof surface

Canonical route: /signal-canvas/peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing

stale

Proof freshness: stale
Proof status: unverified
Display score: 7/10
Last proof check: 2026-03-30
Score updated: 2026-04-02
Score fresh until: 2026-05-02
References: 43
Source count: 3
Coverage: 50%

This page is showing the last landed evidence receipt and score bundle because the latest proof data is outside the freshness window.

Agent Handoff

Canonical ID peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing | Route /signal-canvas/peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing

REST example

curl https://sciencetostartup.com/api/v1/agent-handoff/signal-canvas/peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing

MCP example

{
  "tool": "search_signal_canvas",
  "arguments": {
    "mode": "paper",
    "paper_ref": "peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing",
    "query_text": "Summarize PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing"
  }
}

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing",
  "normalized_query": "2603.26136",
  "route": "/signal-canvas/peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing",
  "paper_ref": "peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}

Evidence Receipt

Route status: building

Claims: 7

References: 43

Proof: Verification pending

Freshness state: computing

Source paper: PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing

PDF: https://arxiv.org/pdf/2603.26136v1

Source count: 3

Coverage: 50%

Last proof check: 2026-03-30T22:23:54.647Z

Signal Canvas receipt window

Watch and verify: PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing

/buildability/peanut-perturbations-by-eigenvalue-alignment-for-attacking-gnns-under-topology-driven-message-passing

Watchwatch

Subject: PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing

Verdict

Watch

Verdict is Watch because viability or proof quality is intermediate and should be re-evaluated before execution.

Preparing verified analysis

GitHub Code Pulse

No public code linked for this paper yet.

Claim map

Strong 7Mixed 0Weak 0

Evidencepartial
we propose PEANUT, a simple, gradient-free, restricted black-box attack that injects virtual nodes to capitalize on this vulnerability.
Implicationpartial
This claim is explicitly stated in the abstract and introduction, defining the core method.
Verificationpartial
partial
Evidencepartial
PEANUT is a injection based attack, which is widely considered to be more practical and realistic scenario than graph modification attacks, where the attacker is able to modify the original graph structure directly.
Implicationpartial
The abstract clearly states the advantage of injection-based attacks over modification attacks.
Verificationpartial
partial
Evidencepartial
Our method works at the inference phase, making it an evasion attack, and is applicable almost immediately, since it does not involve lengthy iterative optimizations or parameter learning, which add computational and time overhead, or training surrogate models, which are susceptible to failure due to differences in model priors and generalization capabilities.
Implicationpartial
The abstract details the operational phase and efficiency of the attack.
Verificationpartial
partial
Evidencepartial
PEANUT also does not require any features on the injected node and consequently demonstrates that GNN performance can be significantly deteriorated even with injected nodes with zeros for features, highlighting the significance of effectively designed connectivity in such attacks.
Implicationpartial
This is a key finding presented in the abstract and reinforced in the introduction.
Verificationpartial
partial
Evidencepartial
Extensive experiments on real-world datasets across three graph tasks demonstrate the effectiveness of our attack despite its simplicity.
Implicationpartial
The abstract concludes with a statement about the effectiveness of the attack based on extensive experiments.
Verificationpartial
partial
Evidencepartial
GNN architectures explicitly utilize the adjacency (or Laplacian) matrix [7, 15, 25] as message-passing operators. This design choice introduces a major architectural vulnerability: perturbations to the adjacency matrix directly alter the propagation pathways through which information flows, affecting the graph’s receptive fields, aggregation neighborhoods, and spectral properties.
Implicationpartial
The introduction clearly explains the architectural vulnerability that PEANUT targets.
Verificationpartial
partial
Evidencepartial
Theorem 1.Given a trained 2-Layer Simple Graph Convolution network MΘ, and graph G (V, A, X), the perturbationS 𝑣 which maximizesL (S 𝑣 )while satisfyingS𝑣 S⊤𝑣F ≤𝚫, is given byS∗𝑣 =𝚫·u 1v,(6)whereu 1 is the dominant eigenvector ofHH ⊤ (H = X𝚯), andvis any vector with unit norm.
Implicationpartial
This is a direct mathematical result presented as Theorem 1.
Verificationpartial
partial

Author intelligence and commercialization panels stay hidden until the proof receipt is verified, cites at least 3 references, includes at least 2 sources, and clears 50% coverage. The paper narrative and citation surfaces remain public while verification is pending.

PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing

Use Signal Canvas as the narrative proof surface

Use this Signal Canvas via API or MCP

Signal Canvas proof surface