{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4OSMGZ2LJPT2DXT3HPI3YJOV7F","short_pith_number":"pith:4OSMGZ2L","schema_version":"1.0","canonical_sha256":"e3a4c3674b4be7a1de7b3bd1bc25d5f96f592d9c84bff04ffc557a225d3adf1a","source":{"kind":"arxiv","id":"2505.06304","version":4},"attestation_state":"computed","paper":{"title":"SRAF: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Chen Zhi, Chunqiang Hu, Maike Li, Meng Han, Wenpeng Xing, Zhebo Wang, Zhenhua Xu","submitted_at":"2025-05-08T03:21:58Z","abstract_excerpt":"The protection of Intellectual Property (IP) for Large Language Models (LLMs) has become a critical concern as model theft and unauthorized commercialization escalate. While adversarial fingerprinting offers a promising black-box solution for ownership verification, existing methods suffer from significant limitations: they are fragile against downstream model modifications, sensitive to system prompt variations, and easily detectable due to high-perplexity input patterns. In this paper, we propose \\textbf{SRAF}, a stealthy and robust adversarial fingerprinting framework. SRAF employs a synerg"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.06304","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-05-08T03:21:58Z","cross_cats_sorted":[],"title_canon_sha256":"a5f55f06466794e0ce96636e9d512aacc3ba56993e4f1f3a4577dbd7f0a14d4c","abstract_canon_sha256":"ac521f404a197183c16d9640c1af48a9ff2a9f127f361fb3ba3d6d963bdcfd06"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:27.849367Z","signature_b64":"9Epbr8x7hpG04BIRsSzhi3i/j7gQwO+XxsN3cEdjghl68pS3WZhwunAmmY4fddDJAvZHFMJkl+OLQebz8bbpBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3a4c3674b4be7a1de7b3bd1bc25d5f96f592d9c84bff04ffc557a225d3adf1a","last_reissued_at":"2026-05-28T01:04:27.848801Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:27.848801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SRAF: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Chen Zhi, Chunqiang Hu, Maike Li, Meng Han, Wenpeng Xing, Zhebo Wang, Zhenhua Xu","submitted_at":"2025-05-08T03:21:58Z","abstract_excerpt":"The protection of Intellectual Property (IP) for Large Language Models (LLMs) has become a critical concern as model theft and unauthorized commercialization escalate. While adversarial fingerprinting offers a promising black-box solution for ownership verification, existing methods suffer from significant limitations: they are fragile against downstream model modifications, sensitive to system prompt variations, and easily detectable due to high-perplexity input patterns. In this paper, we propose \\textbf{SRAF}, a stealthy and robust adversarial fingerprinting framework. SRAF employs a synerg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.06304","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.06304/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.06304","created_at":"2026-05-28T01:04:27.848876+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.06304v4","created_at":"2026-05-28T01:04:27.848876+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.06304","created_at":"2026-05-28T01:04:27.848876+00:00"},{"alias_kind":"pith_short_12","alias_value":"4OSMGZ2LJPT2","created_at":"2026-05-28T01:04:27.848876+00:00"},{"alias_kind":"pith_short_16","alias_value":"4OSMGZ2LJPT2DXT3","created_at":"2026-05-28T01:04:27.848876+00:00"},{"alias_kind":"pith_short_8","alias_value":"4OSMGZ2L","created_at":"2026-05-28T01:04:27.848876+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.11548","citing_title":"Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends","ref_index":162,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F","json":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F.json","graph_json":"https://pith.science/api/pith-number/4OSMGZ2LJPT2DXT3HPI3YJOV7F/graph.json","events_json":"https://pith.science/api/pith-number/4OSMGZ2LJPT2DXT3HPI3YJOV7F/events.json","paper":"https://pith.science/paper/4OSMGZ2L"},"agent_actions":{"view_html":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F","download_json":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F.json","view_paper":"https://pith.science/paper/4OSMGZ2L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.06304&json=true","fetch_graph":"https://pith.science/api/pith-number/4OSMGZ2LJPT2DXT3HPI3YJOV7F/graph.json","fetch_events":"https://pith.science/api/pith-number/4OSMGZ2LJPT2DXT3HPI3YJOV7F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F/action/storage_attestation","attest_author":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F/action/author_attestation","sign_citation":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F/action/citation_signature","submit_replication":"https://pith.science/pith/4OSMGZ2LJPT2DXT3HPI3YJOV7F/action/replication_record"}},"created_at":"2026-05-28T01:04:27.848876+00:00","updated_at":"2026-05-28T01:04:27.848876+00:00"}