{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3HNIKVINV7VQZE7AFWIOB7HU4Q","short_pith_number":"pith:3HNIKVIN","schema_version":"1.0","canonical_sha256":"d9da85550dafeb0c93e02d90e0fcf4e425cd74429f940b2986417e2eed69ab0e","source":{"kind":"arxiv","id":"2605.24541","version":1},"attestation_state":"computed","paper":{"title":"SemanticZip: A Pilot Framework for Lossy Text Compression with LLMs as Semantic Decompressors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR"],"primary_cat":"cs.LG","authors_text":"Natalia Trukhina, Vadim Vashkelis","submitted_at":"2026-05-23T12:14:04Z","abstract_excerpt":"Text compression for large language model (LLM) systems is usually framed as token deletion, retrieval, summarization, or exact reconstruction. We study a more aggressive but explicitly lossy setting: compress text into compact codes that an LLM can expand into task-relevant meaning. We call this setting SemanticZip. Unlike lossless compression, SemanticZip does not require byte-identical reconstruction; unlike ordinary summarization, it treats model-based decompression as part of the codec and evaluates whether task-relevant semantic commitments are recovered.\n  This paper is a pilot framewor"},"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":"2605.24541","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T12:14:04Z","cross_cats_sorted":["cs.AI","cs.CL","cs.IR"],"title_canon_sha256":"6844f158ddba285b890205ab5fbe3305925aa23751bd0863c2659c84b0cfcd0f","abstract_canon_sha256":"aaf897daa7650d05564b2130a127b598aebfa630947011219aa1e90c10dc00d9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:45.447139Z","signature_b64":"AjH9SdjFyXDKnMWQGq02GrwgRdg8pnNlVLaC4excbIvoYATqncWj5vBvTBXvrYlifAd21o9Ahi56gBbFM2tqCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9da85550dafeb0c93e02d90e0fcf4e425cd74429f940b2986417e2eed69ab0e","last_reissued_at":"2026-05-26T01:03:45.446114Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:45.446114Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SemanticZip: A Pilot Framework for Lossy Text Compression with LLMs as Semantic Decompressors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR"],"primary_cat":"cs.LG","authors_text":"Natalia Trukhina, Vadim Vashkelis","submitted_at":"2026-05-23T12:14:04Z","abstract_excerpt":"Text compression for large language model (LLM) systems is usually framed as token deletion, retrieval, summarization, or exact reconstruction. We study a more aggressive but explicitly lossy setting: compress text into compact codes that an LLM can expand into task-relevant meaning. We call this setting SemanticZip. Unlike lossless compression, SemanticZip does not require byte-identical reconstruction; unlike ordinary summarization, it treats model-based decompression as part of the codec and evaluates whether task-relevant semantic commitments are recovered.\n  This paper is a pilot framewor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24541","kind":"arxiv","version":1},"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/2605.24541/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":"2605.24541","created_at":"2026-05-26T01:03:45.446262+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24541v1","created_at":"2026-05-26T01:03:45.446262+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24541","created_at":"2026-05-26T01:03:45.446262+00:00"},{"alias_kind":"pith_short_12","alias_value":"3HNIKVINV7VQ","created_at":"2026-05-26T01:03:45.446262+00:00"},{"alias_kind":"pith_short_16","alias_value":"3HNIKVINV7VQZE7A","created_at":"2026-05-26T01:03:45.446262+00:00"},{"alias_kind":"pith_short_8","alias_value":"3HNIKVIN","created_at":"2026-05-26T01:03:45.446262+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q","json":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q.json","graph_json":"https://pith.science/api/pith-number/3HNIKVINV7VQZE7AFWIOB7HU4Q/graph.json","events_json":"https://pith.science/api/pith-number/3HNIKVINV7VQZE7AFWIOB7HU4Q/events.json","paper":"https://pith.science/paper/3HNIKVIN"},"agent_actions":{"view_html":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q","download_json":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q.json","view_paper":"https://pith.science/paper/3HNIKVIN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24541&json=true","fetch_graph":"https://pith.science/api/pith-number/3HNIKVINV7VQZE7AFWIOB7HU4Q/graph.json","fetch_events":"https://pith.science/api/pith-number/3HNIKVINV7VQZE7AFWIOB7HU4Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q/action/storage_attestation","attest_author":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q/action/author_attestation","sign_citation":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q/action/citation_signature","submit_replication":"https://pith.science/pith/3HNIKVINV7VQZE7AFWIOB7HU4Q/action/replication_record"}},"created_at":"2026-05-26T01:03:45.446262+00:00","updated_at":"2026-05-26T01:03:45.446262+00:00"}