ARXIV:2604.21310 · MALWARE DETECTION · SUBMITTED 24 APR · 20:31 UTC · FRESHNESS STALE

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: unverified proof status

Adversarial Evasion in Non-Stationary Malware Detection: Minimizing Drift Signals through Similarity-Constrained Perturbations

Pawan Acharya · Lan Zhang · arXiv

A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.

Ship in 2-4 weeks›Score4.0Evidence unverified

Opportunity summary

Pain A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.

Evidence 0 refs | 3 sources | 50% coverage

Blocker Evidence unverified

Open Build Read PDF Signal Canvas Track

PROBLEM

A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations. However, these models face critical limitations in real-world, non-stationary environments where both malware characteristics and detection…

METHOD

Full abstract

Deep learning has emerged as a powerful approach for malware detection, demonstrating impressive accuracy across various data representations. However, these models face critical limitations in real-world, non-stationary environments where both malware characteristics and detection systems continuously evolve. Our research investigates a fundamental security question: Can an attacker generate adversarial malware samples that simultaneously evade classification and remain inconspicuous to drift monitoring mechanisms? We propose a novel approach that generates targeted adversarial examples in the classifier's standardized feature space, augmented with sophisticated similarity regularizers. By carefully constraining perturbations to maintain distributional similarity with clean malware, we create an optimization objective that balances targeted misclassification with drift signal minimization. We quantify the effectiveness of this approach by comprehensively comparing classifier output probabilities using multiple drift metrics. Our experiments demonstrate that similarity constraints can reduce output drift signals, with $\ell_2$ regularization showing the most promising results. We observe that perturbation budget significantly influences the evasion-detectability trade-off, with increased budget leading to higher attack success rates and more substantial drift indicators.

RESULT

ScienceToStartup currently rates this 4.0/10 on the public viability pass. Our experiments demonstrate that similarity constraints can reduce output drift signals, with $\ell_2$ regularization showing the most promising results. Code availability is flagged in…

WHY NOW

Malware Detection moved forward this cycle; last verified April 2026. Public score 4.0/10. Production flags indicate code availability.

Continue into Read for claims, analysis, references, and neighboring papers.

Opportunity summary

Score4.0

PainA novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.

Evidence0 refs | 3 sources | 50% coverage

Blockerno shell-level blocker reported

Analysis summary

A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.

VerifiedSource: PDF linkedVerifiedPaperPack: citation fields availablePartialProof: unverified proof status

Competitive landscape

A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.

Segment

Malware Detection

Adoption evidence

No public code link in the paper record yet

Commercial read

4.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

{ "contract_version": "paper-r2", "paper_id": "ba9f580b-ecfd-4bc5-8b4b-27b1a0d0857f", "arxiv_id": "2604.21310", "canonical_route": "/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba", "active_tab": "synced from current hash by the drawer client", "selected_artifact": "adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba", "endpoints": { "paper_pack": "/api/v1/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba/paper-pack", "build_passport": "/api/v1/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba/build-passport", "mcp_resource": "sciencetostartup://surfaces/paper-workspace" } }

{ "surface": "paper", "mode": "paper", "query": "Adversarial Evasion in Non-Stationary Malware Detection: Minimizing Drift Signals through Similarity-Constrained Perturbations", "normalized_query": "2604.21310", "route": "/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba", "paper_ref": "adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba", "topic_slug": null, "benchmark_ref": null, "dataset_ref": null }

{ "@context": "https://schema.org", "@graph": [ { "@type": "WebPage", "@id": "https://sciencetostartup.com/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba#webpage", "url": "https://sciencetostartup.com/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba", "name": "Adversarial Evasion in Non-Stationary Malware Detection: Minimizing Drift Signals through Similarity-Constrained Perturbations", "description": "A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.", "isPartOf": { "@id": "https://sciencetostartup.com/#website" } }, { "@type": "ScholarlyArticle", "@id": "https://sciencetostartup.com/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba#scholarlyArticle", "headline": "Adversarial Evasion in Non-Stationary Malware Detection: Minimizing Drift Signals through Similarity-Constrained Perturbations", "description": "A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.", "url": "https://sciencetostartup.com/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba", "sameAs": "https://arxiv.org/abs/2604.21310", "identifier": { "@type": "PropertyValue", "propertyID": "arXiv", "value": "2604.21310" }, "isAccessibleForFree": true, "isPartOf": { "@id": "https://sciencetostartup.com/#website" }, "datePublished": "2026-04-23T06:03:50.000Z", "author": [ { "@type": "Person", "name": "Pawan Acharya" }, { "@type": "Person", "name": "Lan Zhang" } ], "additionalProperty": [ { "@type": "PropertyValue", "propertyID": "viabilityScore", "value": 4 }, { "@type": "PropertyValue", "propertyID": "researchDomain", "value": "Malware Detection" }, { "@type": "PropertyValue", "propertyID": "commercialReadiness", "value": "code" } ] }, { "@type": "BreadcrumbList", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://sciencetostartup.com" }, { "@type": "ListItem", "position": 2, "name": "Malware Detection", "item": "https://sciencetostartup.com/topics" }, { "@type": "ListItem", "position": 3, "name": "Adversarial Evasion in Non-Stationary Malware Detection: Min", "item": "https://sciencetostartup.com/paper/adversarial-evasion-in-non-stationary-malware-detection-minimizing-drift-signals-through-similarity-constrained-perturba" } ] } ] }

Competitive landscape

A novel approach to generate adversarial malware samples that evade detection and minimize drift signals through similarity-constrained perturbations.

Segment

Malware Detection

Adoption evidence

No public code link in the paper record yet

Commercial read

4.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

Adversarial Evasion in Non-Stationary Malware Detection: Minimizing Drift Signals through Similarity-Constrained Perturbations

Adversarial Evasion in Non-Stationary Malware Detection: Minimizing Drift Signals through Similarity-Constrained Perturbations

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline

Claim map

Constellation map

Competitive landscape

Buzz

PDF

REFERENCES

Related Papers

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline