{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NO6OUTADQNIL2C77HMEUMDEXIP","short_pith_number":"pith:NO6OUTAD","schema_version":"1.0","canonical_sha256":"6bbcea4c038350bd0bff3b09460c9743f1d3a8593ccc741ef5de08126ac16fcb","source":{"kind":"arxiv","id":"2505.11541","version":2},"attestation_state":"computed","paper":{"title":"MorphMark: Flexible Adaptive Watermarking for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Baoyuan Wu, Tianle Gu, Yujiu Yang, Zongqi Wang","submitted_at":"2025-05-14T13:11:16Z","abstract_excerpt":"Watermarking by altering token sampling probabilities based on red-green list is a promising method for tracing the origin of text generated by large language models (LLMs). However, existing watermark methods often struggle with a fundamental dilemma: improving watermark effectiveness (the detectability of the watermark) often comes at the cost of reduced text quality. This trade-off limits their practical application. To address this challenge, we first formalize the problem within a multi-objective trade-off analysis framework. Within this framework, we identify a key factor that influences"},"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":"2505.11541","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-05-14T13:11:16Z","cross_cats_sorted":[],"title_canon_sha256":"bce5a5996a8bc29bd85a94038b3dad99ae203f815cbc2335b0b35dccc8be259b","abstract_canon_sha256":"7d184835d21f1235b2ae378ee423e245f7dc82283e85dc801b490db21f72d160"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:44.913726Z","signature_b64":"xgHPE7ub9j85usE6Cp5t1R2tS8Jc0vbbZPsAxIndE8RGikGJAvHw4W/KXanLU/bL3nkYEZQKM6F372e6M5KHDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6bbcea4c038350bd0bff3b09460c9743f1d3a8593ccc741ef5de08126ac16fcb","last_reissued_at":"2026-07-05T11:05:44.913193Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:44.913193Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MorphMark: Flexible Adaptive Watermarking for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Baoyuan Wu, Tianle Gu, Yujiu Yang, Zongqi Wang","submitted_at":"2025-05-14T13:11:16Z","abstract_excerpt":"Watermarking by altering token sampling probabilities based on red-green list is a promising method for tracing the origin of text generated by large language models (LLMs). However, existing watermark methods often struggle with a fundamental dilemma: improving watermark effectiveness (the detectability of the watermark) often comes at the cost of reduced text quality. This trade-off limits their practical application. To address this challenge, we first formalize the problem within a multi-objective trade-off analysis framework. Within this framework, we identify a key factor that influences"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11541","kind":"arxiv","version":2},"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/2505.11541/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":"2505.11541","created_at":"2026-07-05T11:05:44.913261+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.11541v2","created_at":"2026-07-05T11:05:44.913261+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11541","created_at":"2026-07-05T11:05:44.913261+00:00"},{"alias_kind":"pith_short_12","alias_value":"NO6OUTADQNIL","created_at":"2026-07-05T11:05:44.913261+00:00"},{"alias_kind":"pith_short_16","alias_value":"NO6OUTADQNIL2C77","created_at":"2026-07-05T11:05:44.913261+00:00"},{"alias_kind":"pith_short_8","alias_value":"NO6OUTAD","created_at":"2026-07-05T11:05:44.913261+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25796","citing_title":"SAMark: A Self-Anchored Text Watermarking with Paragraph-Level Paraphrase Robustness","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12456","citing_title":"TextSeal: A Localized LLM Watermark for Provenance & Distillation Protection","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2508.11548","citing_title":"Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends","ref_index":149,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12456","citing_title":"TextSeal: A Localized LLM Watermark for Provenance & Distillation Protection","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05242","citing_title":"XMark: Reliable Multi-Bit Watermarking for LLM-Generated Texts","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP","json":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP.json","graph_json":"https://pith.science/api/pith-number/NO6OUTADQNIL2C77HMEUMDEXIP/graph.json","events_json":"https://pith.science/api/pith-number/NO6OUTADQNIL2C77HMEUMDEXIP/events.json","paper":"https://pith.science/paper/NO6OUTAD"},"agent_actions":{"view_html":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP","download_json":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP.json","view_paper":"https://pith.science/paper/NO6OUTAD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.11541&json=true","fetch_graph":"https://pith.science/api/pith-number/NO6OUTADQNIL2C77HMEUMDEXIP/graph.json","fetch_events":"https://pith.science/api/pith-number/NO6OUTADQNIL2C77HMEUMDEXIP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP/action/storage_attestation","attest_author":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP/action/author_attestation","sign_citation":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP/action/citation_signature","submit_replication":"https://pith.science/pith/NO6OUTADQNIL2C77HMEUMDEXIP/action/replication_record"}},"created_at":"2026-07-05T11:05:44.913261+00:00","updated_at":"2026-07-05T11:05:44.913261+00:00"}