mlsanityData to 14 Oct 2026

KBF: Knowledge Boundary as Fingerprint for Language Model and Black-Box API Auditing

arXiv 2605.29524 ↗First posted 28 May 2026cs.CR cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

Relay and reseller APIs mediate access to large language models (LLMs), but users cannot directly verify which model serves them. We introduce name, a black-box auditing protocol based on stable factual recall near the knowledge boundary, including repeatable wrong answers. KBF generates benign, renewable probes and calibrates audit decisions against reference self-variation. Across 16 production endpoints, KBF detects all 155 economically relevant substitutions without rejecting any of the 16 same-reference controls. KBF remains robust to deployment variation and reaches 95% TPR at a substitution rate as low as 15% in mixed-routing simulations. Field audits flag 7 of 28 endpoints across six platforms as statistically inconsistent with their references. After reference enrollment, even GPT-6 Astra costs only approximately 0.67 per online audit at the recorded API prices.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 34 of its 68 references that are in the map. First posted on arXiv 28 May 2026.

Builds on

34 of 68 references.

References · 34

  1. 2023A Watermark for Large Language Modelscited by 725
  2. 2018Turning Your Weakness Into a Strength: Watermarking Deep Neural Networks by Backdooringcited by 357
  3. 2022Understanding Zero-Shot Adversarial Robustness for Large-Scale Modelscited by 117
  4. 2018Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardwarecited by 102
  5. 2024Instructional Fingerprinting of Large Language Modelscited by 68
  6. 2024LLMmap: Fingerprinting For Large Language Modelscited by 58
  7. 2024zkLLM: Zero Knowledge Proofs for Large Language Modelscited by 51
  8. 2024Model Equality Testing: Which Model Is This API Serving?cited by 47
  9. 2024Knowledge Boundary of Large Language Models: A Surveycited by 40
  10. 2024Hey, That's My Model! Introducing Chain & Hash, An LLM Fingerprinting Techniquecited by 37
  11. 2024TRAP: Targeted Random Adversarial Prompt Honeypot for Black-Box Identificationcited by 36
  12. 2024A Fingerprint for Large Language Modelscited by 28
22 more
  1. 2025Auditing Black-Box LLM APIs with a Rank-Based Uniformity Testcited by 18
  2. 2026Incompressible Knowledge Probes: Estimating Black-Box LLM Parameter Counts via Factual Capacitycited by 17
  3. 2025PMark: Towards Robust and Distortion-free Semantic-level Watermarking with Channel Constraintscited by 16
  4. 2025Gradient-Based Model Fingerprinting for LLM Similarity Detection and Family Classificationcited by 14
  5. 2026Your Agent Is Mine: Measuring Malicious Intermediary Attacks on the LLM Supply Chaincited by 13
  6. 2025EverTracer: Hunting Stolen Large Language Models via Stealthy and Robust Probabilistic Fingerprintcited by 12
  7. 2025Log Probability Tracking of LLM APIscited by 10
  8. 2025TeleSparse: Practical Privacy-Preserving Verification of Deep Neural Networkscited by 9
  9. 2025Fingerprinting LLMs via Prompt Injectioncited by 8
  10. 2025Are Robust LLM Fingerprints Adversarially Robust?cited by 8
  11. 2026Real Money, Fake Models: Deceptive Model Claims in Shadow APIscited by 8
  12. 2025Do Retrieval Augmented Language Models Know When They Don't Know?cited by 5
  13. 2026Token-Efficient Change Detection in LLM APIscited by 5
  14. 2023The Daunting Dilemma with Sentence Encoders: Success on Standard Benchmarks, Failure in Capturing Basic Semantic Propertiescited by 4
  15. 2024I'm Spartacus, No, I'm Spartacus: Measuring and Understanding LLM Identity Confusioncited by 4
  16. 2025FDLLM: A Dedicated Detector for Black-Box LLMs Fingerprintingcited by 4
  17. 2025Every Language Model Has a Forgery-Resistant Signaturecited by 4
  18. 2026Behavioral Fingerprints for LLM Endpoint Stability and Identitycited by 4
  19. 2026AttnDiff: Attention-based Differential Fingerprinting for Large Language Modelscited by 4
  20. 2026IMMACULATE: A Practical LLM Auditing Framework via Verifiable Computationcited by 3
  21. 2026Knowledge Boundary Discovery for Large Language Modelscited by 1
  22. 2026Behavioral Consistency and Transparency Analysis on Large Language Model API Gatewayscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Stemma: Induced Decision Regions Reveal LLM Provenancecited by 0
  2. 2026Ventor-QTest: Threat-Model-Driven Verification of Vendor-Hosted LLM APIscited by 0
  3. 2026Token Counts Are Not Model Lineage: A Frozen-Threshold Holdout Study of Black-Box LLM API Fingerprintingcited by 0